Run tests on this branch

Co-authored-by: Lincoln Stein <lstein@gmail.com>

.github/actions/install-frontend-deps/action.yml

@@ -0,0 +1,33 @@
+name: install frontend dependencies
+description: Installs frontend dependencies with pnpm, with caching
+runs:
+  using: 'composite'
+  steps:
+    - name: setup node 18
+      uses: actions/setup-node@v4
+      with:
+        node-version: '18'
+
+    - name: setup pnpm
+      uses: pnpm/action-setup@v2
+      with:
+        version: 8
+        run_install: false
+
+    - name: get pnpm store directory
+      shell: bash
+      run: |
+        echo "STORE_PATH=$(pnpm store path --silent)" >> $GITHUB_ENV
+
+    - name: setup cache
+      uses: actions/cache@v4
+      with:
+        path: ${{ env.STORE_PATH }}
+        key: ${{ runner.os }}-pnpm-store-${{ hashFiles('**/pnpm-lock.yaml') }}
+        restore-keys: |
+          ${{ runner.os }}-pnpm-store-
+
+    - name: install frontend dependencies
+      run: pnpm install --prefer-frozen-lockfile
+      shell: bash
+      working-directory: invokeai/frontend/web

.github/pr_labels.yml

@@ -1,59 +1,59 @@
-Root:
+root:
 - changed-files:
   - any-glob-to-any-file: '*'
 
-PythonDeps:
+python-deps:
 - changed-files:
   - any-glob-to-any-file: 'pyproject.toml'
 
-Python:
+python:
 - changed-files:
   - all-globs-to-any-file: 
     - 'invokeai/**'
     - '!invokeai/frontend/web/**'
 
-PythonTests:
+python-tests:
 - changed-files:
   - any-glob-to-any-file: 'tests/**'
 
-CICD:
+ci-cd:
 - changed-files:
   - any-glob-to-any-file: .github/**
 
-Docker:
+docker:
 - changed-files:
   - any-glob-to-any-file: docker/**
 
-Installer:
+installer:
 - changed-files:
   - any-glob-to-any-file: installer/**
 
-Documentation:
+docs:
 - changed-files:
   - any-glob-to-any-file: docs/**
 
-Invocations:
+invocations:
 - changed-files:
   - any-glob-to-any-file: 'invokeai/app/invocations/**'
 
-Backend:
+backend:
 - changed-files:
   - any-glob-to-any-file: 'invokeai/backend/**'
 
-Api:
+api:
 - changed-files:
   - any-glob-to-any-file: 'invokeai/app/api/**'
 
-Services:
+services:
 - changed-files:
   - any-glob-to-any-file: 'invokeai/app/services/**'
 
-FrontendDeps:
+frontend-deps:
 - changed-files:
   - any-glob-to-any-file:
     - '**/*/package.json'
     - '**/*/pnpm-lock.yaml'
 
-Frontend:
+frontend:
 - changed-files:
   - any-glob-to-any-file: 'invokeai/frontend/web/**'

.github/workflows/build-container.yml

@@ -11,7 +11,7 @@ on:
       - 'docker/docker-entrypoint.sh'
       - 'workflows/build-container.yml'
     tags:
-      - 'v*'
+      - 'v*.*.*'
   workflow_dispatch:
 
 permissions:

.github/workflows/build-installer.yml

@@ -0,0 +1,45 @@
+# Builds and uploads the installer and python build artifacts.
+
+name: build installer
+
+on:
+  workflow_dispatch:
+  workflow_call:
+
+jobs:
+  build-installer:
+    runs-on: ubuntu-latest
+    timeout-minutes: 5 # expected run time: <2 min
+    steps:
+      - name: checkout
+        uses: actions/checkout@v4
+
+      - name: setup python
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.10'
+          cache: pip
+          cache-dependency-path: pyproject.toml
+
+      - name: install pypa/build
+        run: pip install --upgrade build
+
+      - name: setup frontend
+        uses: ./.github/actions/install-frontend-deps
+
+      - name: create installer
+        id: create_installer
+        run: ./create_installer.sh
+        working-directory: installer
+
+      - name: upload python distribution artifact
+        uses: actions/upload-artifact@v4
+        with:
+          name: dist
+          path: ${{ steps.create_installer.outputs.DIST_PATH }}
+
+      - name: upload installer artifact
+        uses: actions/upload-artifact@v4
+        with:
+          name: ${{ steps.create_installer.outputs.INSTALLER_FILENAME }}
+          path: ${{ steps.create_installer.outputs.INSTALLER_PATH }}

.github/workflows/frontend-checks.yml

@@ -0,0 +1,80 @@
+# Runs frontend code quality checks.
+#
+# Checks for changes to frontend files before running the checks.
+# If always_run is true, always runs the checks.
+
+name: 'frontend checks'
+
+on:
+  push:
+    branches:
+      - 'main'
+  pull_request:
+    types:
+      - 'ready_for_review'
+      - 'opened'
+      - 'synchronize'
+  merge_group:
+  workflow_dispatch:
+    inputs:
+      always_run:
+        description: 'Always run the checks'
+        required: true
+        type: boolean
+        default: true
+  workflow_call:
+    inputs:
+      always_run:
+        description: 'Always run the checks'
+        required: true
+        type: boolean
+        default: true
+
+defaults:
+  run:
+    working-directory: invokeai/frontend/web
+
+jobs:
+  frontend-checks:
+    runs-on: ubuntu-latest
+    timeout-minutes: 10 # expected run time: <2 min
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: check for changed frontend files
+        if: ${{ inputs.always_run != true }}
+        id: changed-files
+        uses: tj-actions/changed-files@v42
+        with:
+          files_yaml: |
+            frontend:
+              - 'invokeai/frontend/web/**'
+
+      - name: install dependencies
+        if: ${{ steps.changed-files.outputs.frontend_any_changed == 'true' || inputs.always_run == true }}
+        uses: ./.github/actions/install-frontend-deps
+
+      - name: tsc
+        if: ${{ steps.changed-files.outputs.frontend_any_changed == 'true' || inputs.always_run == true }}
+        run: 'pnpm lint:tsc'
+        shell: bash
+
+      - name: dpdm
+        if: ${{ steps.changed-files.outputs.frontend_any_changed == 'true' || inputs.always_run == true }}
+        run: 'pnpm lint:dpdm'
+        shell: bash
+
+      - name: eslint
+        if: ${{ steps.changed-files.outputs.frontend_any_changed == 'true' || inputs.always_run == true }}
+        run: 'pnpm lint:eslint'
+        shell: bash
+
+      - name: prettier
+        if: ${{ steps.changed-files.outputs.frontend_any_changed == 'true' || inputs.always_run == true }}
+        run: 'pnpm lint:prettier'
+        shell: bash
+
+      - name: knip
+        if: ${{ steps.changed-files.outputs.frontend_any_changed == 'true' || inputs.always_run == true }}
+        run: 'pnpm lint:knip'
+        shell: bash

.github/workflows/frontend-tests.yml

@@ -0,0 +1,60 @@
+# Runs frontend tests.
+#
+# Checks for changes to frontend files before running the tests.
+# If always_run is true, always runs the tests.
+
+name: 'frontend tests'
+
+on:
+  push:
+    branches:
+      - 'main'
+  pull_request:
+    types:
+      - 'ready_for_review'
+      - 'opened'
+      - 'synchronize'
+  merge_group:
+  workflow_dispatch:
+    inputs:
+      always_run:
+        description: 'Always run the tests'
+        required: true
+        type: boolean
+        default: true
+  workflow_call:
+    inputs:
+      always_run:
+        description: 'Always run the tests'
+        required: true
+        type: boolean
+        default: true
+
+defaults:
+  run:
+    working-directory: invokeai/frontend/web
+
+jobs:
+  frontend-tests:
+    runs-on: ubuntu-latest
+    timeout-minutes: 10 # expected run time: <2 min
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: check for changed frontend files
+        if: ${{ inputs.always_run != true }}
+        id: changed-files
+        uses: tj-actions/changed-files@v42
+        with:
+          files_yaml: |
+            frontend:
+              - 'invokeai/frontend/web/**'
+
+      - name: install dependencies
+        if: ${{ steps.changed-files.outputs.frontend_any_changed == 'true' || inputs.always_run == true }}
+        uses: ./.github/actions/install-frontend-deps
+
+      - name: vitest
+        if: ${{ steps.changed-files.outputs.frontend_any_changed == 'true' || inputs.always_run == true }}
+        run: 'pnpm test:no-watch'
+        shell: bash

.github/workflows/label-pr.yml

@@ -1,6 +1,6 @@
-name: "Pull Request Labeler"
+name: 'label PRs'
 on:
-- pull_request_target
+  - pull_request_target
 
 jobs:
   labeler:
@@ -9,8 +9,10 @@ jobs:
       pull-requests: write
     runs-on: ubuntu-latest
     steps:
-      - name: Checkout
+      - name: checkout
         uses: actions/checkout@v4
-      - uses: actions/labeler@v5
+
+      - name: label PRs
+        uses: actions/labeler@v5
         with:
           configuration-path: .github/pr_labels.yml

.github/workflows/lint-frontend.yml

@@ -1,45 +0,0 @@
-name: Lint frontend
-
-on:
-  pull_request:
-    types:
-      - 'ready_for_review'
-      - 'opened'
-      - 'synchronize'
-  push:
-    branches:
-      - 'main'
-  merge_group:
-  workflow_dispatch:
-
-defaults:
-  run:
-    working-directory: invokeai/frontend/web
-
-jobs:
-  lint-frontend:
-    if: github.event.pull_request.draft == false
-    runs-on: ubuntu-22.04
-    steps:
-      - name: Setup Node 18
-        uses: actions/setup-node@v4
-        with:
-          node-version: '18'
-      - name: Checkout
-        uses: actions/checkout@v4
-      - name: Setup pnpm
-        uses: pnpm/action-setup@v2
-        with:
-          version: '8.12.1'
-      - name: Install dependencies
-        run: 'pnpm install --prefer-frozen-lockfile'
-      - name: Typescript
-        run: 'pnpm run lint:tsc'
-      - name: Madge
-        run: 'pnpm run lint:dpdm'
-      - name: ESLint
-        run: 'pnpm run lint:eslint'
-      - name: Prettier
-        run: 'pnpm run lint:prettier'
-      - name: Knip
-        run: 'pnpm run lint:knip'

.github/workflows/mkdocs-material.yml

@@ -1,51 +1,49 @@
-name: mkdocs-material
+# This is a mostly a copy-paste from https://github.com/squidfunk/mkdocs-material/blob/master/docs/publishing-your-site.md
+
+name: mkdocs
+
 on:
   push:
     branches:
-      - 'refs/heads/main'
+      - main
+  workflow_dispatch:
 
 permissions:
   contents: write
 
 jobs:
-  mkdocs-material:
+  deploy:
     if: github.event.pull_request.draft == false
     runs-on: ubuntu-latest
     env:
       REPO_URL: '${{ github.server_url }}/${{ github.repository }}'
       REPO_NAME: '${{ github.repository }}'
       SITE_URL: 'https://${{ github.repository_owner }}.github.io/InvokeAI'
+
     steps:
-      - name: checkout sources
-        uses: actions/checkout@v3
-        with:
-          fetch-depth: 0
+      - name: checkout
+        uses: actions/checkout@v4
 
       - name: setup python
-        uses: actions/setup-python@v4
+        uses: actions/setup-python@v5
         with:
           python-version: '3.10'
           cache: pip
           cache-dependency-path: pyproject.toml
 
-      - name: install requirements
-        env:
-          PIP_USE_PEP517: 1
-        run: |
-          python -m \
-            pip install ".[docs]"
+      - name: set cache id
+        run: echo "cache_id=$(date --utc '+%V')" >> $GITHUB_ENV
 
-      - name: confirm buildability
-        run: |
-          python -m \
-            mkdocs build \
-            --clean \
-            --verbose
+      - name: use cache
+        uses: actions/cache@v4
+        with:
+          key: mkdocs-material-${{ env.cache_id }}
+          path: .cache
+          restore-keys: |
+            mkdocs-material-
 
-      - name: deploy to gh-pages
-        if: ${{ github.ref == 'refs/heads/main' }}
-        run: |
-          python -m \
-            mkdocs gh-deploy \
-            --clean \
-            --force
+      - name: install dependencies
+        run: python -m pip install ".[docs]"
+
+      - name: build & deploy
+        run: mkdocs gh-deploy --force

.github/workflows/pypi-release.yml

@@ -1,67 +0,0 @@
-name: PyPI Release
-
-on:
-  workflow_dispatch:
-    inputs:
-      publish_package:
-        description: 'Publish build on PyPi? [true/false]'
-        required: true
-        default: 'false'
-
-jobs:
-  build-and-release:
-    if: github.repository == 'invoke-ai/InvokeAI'
-    runs-on: ubuntu-22.04
-    env:
-      TWINE_USERNAME: __token__
-      TWINE_PASSWORD: ${{ secrets.PYPI_API_TOKEN }}
-      TWINE_NON_INTERACTIVE: 1
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-
-      - name: Setup Node 18
-        uses: actions/setup-node@v4
-        with:
-          node-version: '18'
-
-      - name: Setup pnpm
-        uses: pnpm/action-setup@v2
-        with:
-          version: '8.12.1'
-
-      - name: Install frontend dependencies
-        run: pnpm install --prefer-frozen-lockfile
-        working-directory: invokeai/frontend/web
-
-      - name: Build frontend
-        run: pnpm run build
-        working-directory: invokeai/frontend/web
-
-      - name: Install python dependencies
-        run: pip install --upgrade build twine
-
-      - name: Build python package
-        run: python3 -m build
-
-      - name: Upload build as workflow artifact
-        uses: actions/upload-artifact@v4
-        with:
-          name: dist
-          path: dist
-
-      - name: Check distribution
-        run: twine check dist/*
-
-      - name: Check PyPI versions
-        if: github.ref == 'refs/heads/main' || startsWith(github.ref, 'refs/heads/release/')
-        run: |
-          pip install --upgrade requests
-          python -c "\
-          import scripts.pypi_helper; \
-          EXISTS=scripts.pypi_helper.local_on_pypi(); \
-          print(f'PACKAGE_EXISTS={EXISTS}')" >> $GITHUB_ENV
-
-      - name: Publish build on PyPi
-        if: env.PACKAGE_EXISTS == 'False' && env.TWINE_PASSWORD != '' && github.event.inputs.publish_package == 'true'
-        run: twine upload dist/*

.github/workflows/python-checks.yml

@@ -0,0 +1,76 @@
+# Runs python code quality checks.
+#
+# Checks for changes to python files before running the checks.
+# If always_run is true, always runs the checks.
+#
+# TODO: Add mypy or pyright to the checks.
+
+name: 'python checks'
+
+on:
+  push:
+    branches:
+      - 'main'
+  pull_request:
+    types:
+      - 'ready_for_review'
+      - 'opened'
+      - 'synchronize'
+  merge_group:
+  workflow_dispatch:
+    inputs:
+      always_run:
+        description: 'Always run the checks'
+        required: true
+        type: boolean
+        default: true
+  workflow_call:
+    inputs:
+      always_run:
+        description: 'Always run the checks'
+        required: true
+        type: boolean
+        default: true
+
+jobs:
+  python-checks:
+    runs-on: ubuntu-latest
+    timeout-minutes: 5 # expected run time: <1 min
+    steps:
+      - name: checkout
+        uses: actions/checkout@v4
+
+      - name: check for changed python files
+        if: ${{ inputs.always_run != true }}
+        id: changed-files
+        uses: tj-actions/changed-files@v42
+        with:
+          files_yaml: |
+            python:
+              - 'pyproject.toml'
+              - 'invokeai/**'
+              - '!invokeai/frontend/web/**'
+              - 'tests/**'
+
+      - name: setup python
+        if: ${{ steps.changed-files.outputs.python_any_changed == 'true' || inputs.always_run == true }}
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.10'
+          cache: pip
+          cache-dependency-path: pyproject.toml
+
+      - name: install ruff
+        if: ${{ steps.changed-files.outputs.python_any_changed == 'true' || inputs.always_run == true }}
+        run: pip install ruff
+        shell: bash
+
+      - name: ruff check
+        if: ${{ steps.changed-files.outputs.python_any_changed == 'true' || inputs.always_run == true }}
+        run: ruff check --output-format=github .
+        shell: bash
+
+      - name: ruff format
+        if: ${{ steps.changed-files.outputs.python_any_changed == 'true' || inputs.always_run == true }}
+        run: ruff format --check .
+        shell: bash

.github/workflows/python-tests.yml

@@ -0,0 +1,107 @@
+# Runs python tests on a matrix of python versions and platforms.
+#
+# Checks for changes to python files before running the tests.
+# If always_run is true, always runs the tests.
+
+name: 'python tests'
+
+on:
+  push:
+    branches:
+      - 'main'
+      - 'bug-install-job-running-multiple-times'
+  pull_request:
+    types:
+      - 'ready_for_review'
+      - 'opened'
+      - 'synchronize'
+  merge_group:
+  workflow_dispatch:
+    inputs:
+      always_run:
+        description: 'Always run the tests'
+        required: true
+        type: boolean
+        default: true
+  workflow_call:
+    inputs:
+      always_run:
+        description: 'Always run the tests'
+        required: true
+        type: boolean
+        default: true
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
+  cancel-in-progress: true
+
+jobs:
+  matrix:
+    strategy:
+      matrix:
+        python-version:
+          - '3.10'
+          - '3.11'
+        platform:
+          - linux-cuda-11_7
+          - linux-rocm-5_2
+          - linux-cpu
+          - macos-default
+          - windows-cpu
+        include:
+          - platform: linux-cuda-11_7
+            os: ubuntu-22.04
+            github-env: $GITHUB_ENV
+          - platform: linux-rocm-5_2
+            os: ubuntu-22.04
+            extra-index-url: 'https://download.pytorch.org/whl/rocm5.2'
+            github-env: $GITHUB_ENV
+          - platform: linux-cpu
+            os: ubuntu-22.04
+            extra-index-url: 'https://download.pytorch.org/whl/cpu'
+            github-env: $GITHUB_ENV
+          - platform: macos-default
+            os: macOS-12
+            github-env: $GITHUB_ENV
+          - platform: windows-cpu
+            os: windows-2022
+            github-env: $env:GITHUB_ENV
+    name: 'py${{ matrix.python-version }}: ${{ matrix.platform }}'
+    runs-on: ${{ matrix.os }}
+    timeout-minutes: 15 # expected run time: 2-6 min, depending on platform
+    env:
+      PIP_USE_PEP517: '1'
+    steps:
+      - name: checkout
+        uses: actions/checkout@v4
+
+      - name: check for changed python files
+        if: ${{ inputs.always_run != true }}
+        id: changed-files
+        uses: tj-actions/changed-files@v42
+        with:
+          files_yaml: |
+            python:
+              - 'pyproject.toml'
+              - 'invokeai/**'
+              - '!invokeai/frontend/web/**'
+              - 'tests/**'
+
+      - name: setup python
+        if: ${{ steps.changed-files.outputs.python_any_changed == 'true' || inputs.always_run == true }}
+        uses: actions/setup-python@v5
+        with:
+          python-version: ${{ matrix.python-version }}
+          cache: pip
+          cache-dependency-path: pyproject.toml
+
+      - name: install dependencies
+        if: ${{ steps.changed-files.outputs.python_any_changed == 'true' || inputs.always_run == true }}
+        env:
+          PIP_EXTRA_INDEX_URL: ${{ matrix.extra-index-url }}
+        run: >
+          pip3 install --editable=".[test]"
+
+      - name: run pytest
+        if: ${{ steps.changed-files.outputs.python_any_changed == 'true' || inputs.always_run == true }}
+        run: pytest

.github/workflows/release.yml

@@ -0,0 +1,108 @@
+# Main release workflow. Triggered on tag push or manual trigger.
+#
+# - Runs all code checks and tests
+# - Verifies the app version matches the tag version.
+# - Builds the installer and build, uploading them as artifacts.
+# - Publishes to TestPyPI and PyPI. Both are conditional on the previous steps passing and require a manual approval.
+#
+# See docs/RELEASE.md for more information on the release process.
+
+name: release
+
+on:
+  push:
+    tags:
+      - 'v*'
+  workflow_dispatch:
+
+jobs:
+  check-version:
+    runs-on: ubuntu-latest
+    steps:
+      - name: checkout
+        uses: actions/checkout@v4
+
+      - name: check python version
+        uses: samuelcolvin/check-python-version@v4
+        id: check-python-version
+        with:
+          version_file_path: invokeai/version/invokeai_version.py
+
+  frontend-checks:
+    uses: ./.github/workflows/frontend-checks.yml
+    with:
+      always_run: true
+
+  frontend-tests:
+    uses: ./.github/workflows/frontend-tests.yml
+    with:
+      always_run: true
+
+  python-checks:
+    uses: ./.github/workflows/python-checks.yml
+    with:
+      always_run: true
+
+  python-tests:
+    uses: ./.github/workflows/python-tests.yml
+    with:
+      always_run: true
+
+  build:
+    uses: ./.github/workflows/build-installer.yml
+
+  publish-testpypi:
+    runs-on: ubuntu-latest
+    timeout-minutes: 5 # expected run time: <1 min
+    needs:
+      [
+        check-version,
+        frontend-checks,
+        frontend-tests,
+        python-checks,
+        python-tests,
+        build,
+      ]
+    environment:
+      name: testpypi
+      url: https://test.pypi.org/p/invokeai
+    permissions:
+      id-token: write
+    steps:
+      - name: download distribution from build job
+        uses: actions/download-artifact@v4
+        with:
+          name: dist
+          path: dist/
+
+      - name: publish distribution to TestPyPI
+        uses: pypa/gh-action-pypi-publish@release/v1
+        with:
+          repository-url: https://test.pypi.org/legacy/
+
+  publish-pypi:
+    runs-on: ubuntu-latest
+    timeout-minutes: 5 # expected run time: <1 min
+    needs:
+      [
+        check-version,
+        frontend-checks,
+        frontend-tests,
+        python-checks,
+        python-tests,
+        build,
+      ]
+    environment:
+      name: pypi
+      url: https://pypi.org/p/invokeai
+    permissions:
+      id-token: write
+    steps:
+      - name: download distribution from build job
+        uses: actions/download-artifact@v4
+        with:
+          name: dist
+          path: dist/
+
+      - name: publish distribution to PyPI
+        uses: pypa/gh-action-pypi-publish@release/v1

.github/workflows/style-checks.yml

@@ -1,24 +0,0 @@
-name: style checks
-
-on:
-  pull_request:
-  push:
-    branches: main
-
-jobs:
-  ruff:
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v3
-
-      - name: Setup Python
-        uses: actions/setup-python@v4
-        with:
-          python-version: '3.10'
-
-      - name: Install dependencies with pip
-        run: |
-          pip install ruff
-
-      - run: ruff check --output-format=github .
-      - run: ruff format --check .

.github/workflows/test-invoke-pip.yml

@@ -1,129 +0,0 @@
-name: Test invoke.py pip
-on:
-  push:
-    branches:
-      - 'main'
-  pull_request:
-    types:
-      - 'ready_for_review'
-      - 'opened'
-      - 'synchronize'
-  merge_group:
-  workflow_dispatch:
-
-concurrency:
-  group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
-  cancel-in-progress: true
-
-jobs:
-  matrix:
-    if: github.event.pull_request.draft == false
-    strategy:
-      matrix:
-        python-version:
-          # - '3.9'
-          - '3.10'
-        pytorch:
-          - linux-cuda-11_7
-          - linux-rocm-5_2
-          - linux-cpu
-          - macos-default
-          - windows-cpu
-        include:
-          - pytorch: linux-cuda-11_7
-            os: ubuntu-22.04
-            github-env: $GITHUB_ENV
-          - pytorch: linux-rocm-5_2
-            os: ubuntu-22.04
-            extra-index-url: 'https://download.pytorch.org/whl/rocm5.2'
-            github-env: $GITHUB_ENV
-          - pytorch: linux-cpu
-            os: ubuntu-22.04
-            extra-index-url: 'https://download.pytorch.org/whl/cpu'
-            github-env: $GITHUB_ENV
-          - pytorch: macos-default
-            os: macOS-12
-            github-env: $GITHUB_ENV
-          - pytorch: windows-cpu
-            os: windows-2022
-            github-env: $env:GITHUB_ENV
-    name: ${{ matrix.pytorch }} on ${{ matrix.python-version }}
-    runs-on: ${{ matrix.os }}
-    env:
-      PIP_USE_PEP517: '1'
-    steps:
-      - name: Checkout sources
-        id: checkout-sources
-        uses: actions/checkout@v3
-
-      - name: Check for changed python files
-        id: changed-files
-        uses: tj-actions/changed-files@v41
-        with:
-          files_yaml: |
-            python:
-              - 'pyproject.toml'
-              - 'invokeai/**'
-              - '!invokeai/frontend/web/**'
-              - 'tests/**'
-
-      - name: set test prompt to main branch validation
-        if: steps.changed-files.outputs.python_any_changed == 'true'
-        run: echo "TEST_PROMPTS=tests/validate_pr_prompt.txt" >> ${{ matrix.github-env }}
-
-      - name: setup python
-        if: steps.changed-files.outputs.python_any_changed == 'true'
-        uses: actions/setup-python@v4
-        with:
-          python-version: ${{ matrix.python-version }}
-          cache: pip
-          cache-dependency-path: pyproject.toml
-
-      - name: install invokeai
-        if: steps.changed-files.outputs.python_any_changed == 'true'
-        env:
-          PIP_EXTRA_INDEX_URL: ${{ matrix.extra-index-url }}
-        run: >
-          pip3 install
-          --editable=".[test]"
-
-      - name: run pytest
-        if: steps.changed-files.outputs.python_any_changed == 'true'
-        id: run-pytest
-        run: pytest
-
-      # - name: run invokeai-configure
-      #   env:
-      #     HUGGING_FACE_HUB_TOKEN: ${{ secrets.HUGGINGFACE_TOKEN }}
-      #   run: >
-      #     invokeai-configure
-      #     --yes
-      #     --default_only
-      #     --full-precision
-      #   # can't use fp16 weights without a GPU
-
-      # - name: run invokeai
-      #   id: run-invokeai
-      #   env:
-      #     # Set offline mode to make sure configure preloaded successfully.
-      #     HF_HUB_OFFLINE: 1
-      #     HF_DATASETS_OFFLINE: 1
-      #     TRANSFORMERS_OFFLINE: 1
-      #     INVOKEAI_OUTDIR: ${{ github.workspace }}/results
-      #   run: >
-      #     invokeai
-      #     --no-patchmatch
-      #     --no-nsfw_checker
-      #     --precision=float32
-      #     --always_use_cpu
-      #     --use_memory_db
-      #     --outdir ${{ env.INVOKEAI_OUTDIR }}/${{ matrix.python-version }}/${{ matrix.pytorch }}
-      #     --from_file ${{ env.TEST_PROMPTS }}
-
-      # - name: Archive results
-      #   env:
-      #     INVOKEAI_OUTDIR: ${{ github.workspace }}/results
-      #   uses: actions/upload-artifact@v3
-      #   with:
-      #     name: results
-      #     path: ${{ env.INVOKEAI_OUTDIR }}

.prettierrc.yaml

@@ -7,7 +7,7 @@ embeddedLanguageFormatting: auto
 overrides:
   - files: '*.md'
     options:
-      proseWrap: always
+      proseWrap: preserve
       printWidth: 80
       parser: markdown
       cursorOffset: -1

Makefile

@@ -10,10 +10,12 @@ help:
 	@echo "ruff-unsafe              Run ruff, fixing all fixable errors and formatting"
 	@echo "mypy                     Run mypy using the config in pyproject.toml to identify type mismatches and other coding errors"
 	@echo "mypy-all                 Run mypy ignoring the config in pyproject.tom but still ignoring missing imports"
-	@echo "test"             Run the unit tests.
-	@echo "frontend-install" Install the pnpm modules needed for the front end
+	@echo "test                     Run the unit tests."
+	@echo "update-config-docstring  Update the app's config docstring so mkdocs can autogenerate it correctly."
+	@echo "frontend-install         Install the pnpm modules needed for the front end"
 	@echo "frontend-build           Build the frontend in order to run on localhost:9090"
 	@echo "frontend-dev             Run the frontend in developer mode on localhost:5173"
+	@echo "frontend-typegen         Generate types for the frontend from the OpenAPI schema"
 	@echo "installer-zip            Build the installer .zip file for the current version"
 	@echo "tag-release              Tag the GitHub repository with the current version (use at release time only!)"
 
@@ -40,6 +42,10 @@ mypy-all:
 test:
 	pytest ./tests
 
+# Update config docstring
+update-config-docstring:
+	python scripts/update_config_docstring.py
+
 # Install the pnpm modules needed for the front end
 frontend-install:
 	rm -rf invokeai/frontend/web/node_modules
@@ -53,6 +59,9 @@ frontend-build:
 frontend-dev:
 	cd invokeai/frontend/web && pnpm dev
 
+frontend-typegen:
+	cd invokeai/frontend/web && python ../../../scripts/generate_openapi_schema.py | pnpm typegen
+
 # Installer zip file
 installer-zip:
 	cd installer && ./create_installer.sh

docs/RELEASE.md

@@ -0,0 +1,142 @@
+# Release Process
+
+The app is published in twice, in different build formats.
+
+- A [PyPI] distribution. This includes both a source distribution and built distribution (a wheel). Users install with `pip install invokeai`. The updater uses this build.
+- An installer on the [InvokeAI Releases Page]. This is a zip file with install scripts and a wheel. This is only used for new installs.
+
+## General Prep
+
+Make a developer call-out for PRs to merge. Merge and test things out.
+
+While the release workflow does not include end-to-end tests, it does pause before publishing so you can download and test the final build.
+
+## Release Workflow
+
+The `release.yml` workflow runs a number of jobs to handle code checks, tests, build and publish on PyPI.
+
+It is triggered on **tag push**, when the tag matches `v*`. It doesn't matter if you've prepped a release branch like `release/v3.5.0` or are releasing from `main` - it works the same.
+
+> Because commits are reference-counted, it is safe to create a release branch, tag it, let the workflow run, then delete the branch. So long as the tag exists, that commit will exist.
+
+### Triggering the Workflow
+
+Run `make tag-release` to tag the current commit and kick off the workflow.
+
+The release may also be dispatched [manually].
+
+### Workflow Jobs and Process
+
+The workflow consists of a number of concurrently-run jobs, and two final publish jobs.
+
+The publish jobs require manual approval and are only run if the other jobs succeed.
+
+#### `check-version` Job
+
+This job checks that the git ref matches the app version. It matches the ref against the `__version__` variable in `invokeai/version/invokeai_version.py`.
+
+When the workflow is triggered by tag push, the ref is the tag. If the workflow is run manually, the ref is the target selected from the **Use workflow from** dropdown.
+
+This job uses [samuelcolvin/check-python-version].
+
+> Any valid [version specifier] works, so long as the tag matches the version. The release workflow works exactly the same for `RC`, `post`, `dev`, etc.
+
+#### Check and Test Jobs
+
+- **`python-tests`**: runs `pytest` on matrix of platforms
+- **`python-checks`**: runs `ruff` (format and lint)
+- **`frontend-tests`**: runs `vitest`
+- **`frontend-checks`**: runs `prettier` (format), `eslint` (lint), `dpdm` (circular refs), `tsc` (static type check) and `knip` (unused imports)
+
+> **TODO** We should add `mypy` or `pyright` to the **`check-python`** job.
+
+> **TODO** We should add an end-to-end test job that generates an image.
+
+#### `build-installer` Job
+
+This sets up both python and frontend dependencies and builds the python package. Internally, this runs `installer/create_installer.sh` and uploads two artifacts:
+
+- **`dist`**: the python distribution, to be published on PyPI
+- **`InvokeAI-installer-${VERSION}.zip`**: the installer to be included in the GitHub release
+
+#### Sanity Check & Smoke Test
+
+At this point, the release workflow pauses as the remaining publish jobs require approval.
+
+A maintainer should go to the **Summary** tab of the workflow, download the installer and test it. Ensure the app loads and generates.
+
+> The same wheel file is bundled in the installer and in the `dist` artifact, which is uploaded to PyPI. You should end up with the exactly the same installation of the `invokeai` package from any of these methods.
+
+#### PyPI Publish Jobs
+
+The publish jobs will run if any of the previous jobs fail.
+
+They use [GitHub environments], which are configured as [trusted publishers] on PyPI.
+
+Both jobs require a maintainer to approve them from the workflow's **Summary** tab.
+
+- Click the **Review deployments** button
+- Select the environment (either `testpypi` or `pypi`)
+- Click **Approve and deploy**
+
+> **If the version already exists on PyPI, the publish jobs will fail.** PyPI only allows a given version to be published once - you cannot change it. If version published on PyPI has a problem, you'll need to "fail forward" by bumping the app version and publishing a followup release.
+
+#### `publish-testpypi` Job
+
+Publishes the distribution on the [Test PyPI] index, using the `testpypi` GitHub environment.
+
+This job is not required for the production PyPI publish, but included just in case you want to test the PyPI release.
+
+If approved and successful, you could try out the test release like this:
+
+```sh
+# Create a new virtual environment
+python -m venv ~/.test-invokeai-dist --prompt test-invokeai-dist
+# Install the distribution from Test PyPI
+pip install --index-url https://test.pypi.org/simple/ invokeai
+# Run and test the app
+invokeai-web
+# Cleanup
+deactivate
+rm -rf ~/.test-invokeai-dist
+```
+
+#### `publish-pypi` Job
+
+Publishes the distribution on the production PyPI index, using the `pypi` GitHub environment.
+
+## Publish the GitHub Release with installer
+
+Once the release is published to PyPI, it's time to publish the GitHub release.
+
+1. [Draft a new release] on GitHub, choosing the tag that triggered the release.
+2. Write the release notes, describing important changes. The **Generate release notes** button automatically inserts the changelog and new contributors, and you can copy/paste the intro from previous releases.
+3. Upload the zip file created in **`build`** job into the Assets section of the release notes. You can also upload the zip into the body of the release notes, since it can be hard for users to find the Assets section.
+4. Check the **Set as a pre-release** and **Create a discussion for this release** checkboxes at the bottom of the release page.
+5. Publish the pre-release.
+6. Announce the pre-release in Discord.
+
+> **TODO** Workflows can create a GitHub release from a template and upload release assets. One popular action to handle this is [ncipollo/release-action]. A future enhancement to the release process could set this up.
+
+## Manual Build
+
+The `build installer` workflow can be dispatched manually. This is useful to test the installer for a given branch or tag.
+
+No checks are run, it just builds.
+
+## Manual Release
+
+The `release` workflow can be dispatched manually. You must dispatch the workflow from the right tag, else it will fail the version check.
+
+This functionality is available as a fallback in case something goes wonky. Typically, releases should be triggered via tag push as described above.
+
+[InvokeAI Releases Page]: https://github.com/invoke-ai/InvokeAI/releases
+[PyPI]: https://pypi.org/
+[Draft a new release]: https://github.com/invoke-ai/InvokeAI/releases/new
+[Test PyPI]: https://test.pypi.org/
+[version specifier]: https://packaging.python.org/en/latest/specifications/version-specifiers/
+[ncipollo/release-action]: https://github.com/ncipollo/release-action
+[GitHub environments]: https://docs.github.com/en/actions/deployment/targeting-different-environments/using-environments-for-deployment
+[trusted publishers]: https://docs.pypi.org/trusted-publishers/
+[samuelcolvin/check-python-version]: https://github.com/samuelcolvin/check-python-version
+[manually]: #manual-release

docs/contributing/MODEL_MANAGER.md

@@ -16,11 +16,6 @@ model. These are the:
   information. It is also responsible for managing the InvokeAI
   `models` directory and its contents.
   
-* _ModelMetadataStore_ and _ModelMetaDataFetch_ Backend modules that
-  are able to retrieve metadata from online model repositories,
-  transform them into Pydantic models, and cache them to the InvokeAI
-  SQL database.
-  
 * _DownloadQueueServiceBase_
   A multithreaded downloader responsible
   for downloading models from a remote source to disk. The download
@@ -32,7 +27,6 @@ model. These are the:
   Responsible for loading a model from disk
   into RAM and VRAM and getting it ready for inference.
 
-  
 ## Location of the Code
 
 The four main services can be found in
@@ -67,19 +61,17 @@ provides the following fields:
 | `model_format`   | ModelFormat   | The format of the model (e.g. "diffusers"); also used as a Union discriminator |
 | `base_model`     | BaseModelType | The base model that the model is compatible with |
 | `path`           | str           | Location of model on disk |
-| `original_hash`  | str           | Hash of the model when it was first installed |
-| `current_hash`   | str           | Most recent hash of the model's contents |
+| `hash`           | str           | Hash of the model |
 | `description`    | str           | Human-readable description of the model (optional) |
 | `source`         | str           | Model's source URL or repo id (optional) |
 
 The `key` is a unique 32-character random ID which was generated at
-install time. The `original_hash` field stores a hash of the model's
+install time. The `hash` field stores a hash of the model's
 contents at install time obtained by sampling several parts of the
 model's files using the `imohash` library. Over the course of the
 model's lifetime it may be transformed in various ways, such as
 changing its precision or converting it from a .safetensors to a
-diffusers model. When this happens, `original_hash` is unchanged, but
-`current_hash` is updated to indicate the current contents.
+diffusers model.
 
 `ModelType`, `ModelFormat` and `BaseModelType` are string enums that
 are defined in `invokeai.backend.model_manager.config`. They are also
@@ -94,7 +86,6 @@ The `path` field can be absolute or relative. If relative, it is taken
 to be relative to the `models_dir` setting in the user's
 `invokeai.yaml` file.
 
-
 ### CheckpointConfig
 
 This adds support for checkpoint configurations, and adds the
@@ -228,9 +219,9 @@ The way it works is as follows:
 1. Retrieve the value of the `model_config_db` option from the user's
  `invokeai.yaml` config file.
 2. If `model_config_db` is `auto` (the default), then:
-   - Use the values of `conn` and `lock` to return a `ModelRecordServiceSQL` object
+   * Use the values of `conn` and `lock` to return a `ModelRecordServiceSQL` object
   opened on the passed connection and lock.
-   - Open up a new connection to `databases/invokeai.db` if `conn`
+   * Open up a new connection to `databases/invokeai.db` if `conn`
      and/or `lock` are missing (see note below).
 3. If `model_config_db` is a Path, then use `from_db_file`
    to return the appropriate type of ModelRecordService.
@@ -255,7 +246,7 @@ store = ModelRecordServiceBase.open(config, db_conn, lock)
 
 Configurations can be retrieved in several ways.
 
-#### get_model(key) -> AnyModelConfig:
+#### get_model(key) -> AnyModelConfig
 
 The basic functionality is to call the record store object's
 `get_model()` method with the desired model's unique key. It returns
@@ -272,28 +263,28 @@ print(model_conf.path)
 If the key is unrecognized, this call raises an
 `UnknownModelException`.
 
-#### exists(key) -> AnyModelConfig:
+#### exists(key) -> AnyModelConfig
 
 Returns True if a model with the given key exists in the databsae.
 
-#### search_by_path(path) -> AnyModelConfig:
+#### search_by_path(path) -> AnyModelConfig
 
 Returns the configuration of the model whose path is `path`. The path
 is matched using a simple string comparison and won't correctly match
 models referred to by different paths (e.g. using symbolic links).
 
-#### search_by_name(name, base, type) -> List[AnyModelConfig]:
+#### search_by_name(name, base, type) -> List[AnyModelConfig]
 
 This method searches for models that match some combination of `name`,
 `BaseType` and `ModelType`. Calling without any arguments will return
 all the models in the database.
 
-#### all_models() -> List[AnyModelConfig]:
+#### all_models() -> List[AnyModelConfig]
 
 Return all the model configs in the database. Exactly equivalent to
 calling `search_by_name()` with no arguments.
 
-#### search_by_tag(tags) -> List[AnyModelConfig]:
+#### search_by_tag(tags) -> List[AnyModelConfig]
 
 `tags` is a list of strings. This method returns a list of model
 configs that contain all of the given tags. Examples:
@@ -312,11 +303,11 @@ commercializable_models = [x for x in store.all_models() \
                            if x.license.contains('allowCommercialUse=Sell')]
 ```
 
-#### version() -> str:
+#### version() -> str
 
 Returns the version of the database, currently at `3.2`
 
-#### model_info_by_name(name, base_model, model_type) -> ModelConfigBase:
+#### model_info_by_name(name, base_model, model_type) -> ModelConfigBase
 
 This method exists to ease the transition from the previous version of
 the model manager, in which `get_model()` took the three arguments
@@ -337,7 +328,7 @@ model and pass its key to `get_model()`.
 Several methods allow you to create and update stored model config
 records.
 
-#### add_model(key, config) -> AnyModelConfig:
+#### add_model(key, config) -> AnyModelConfig
 
 Given a key and a configuration, this will add the model's
 configuration record to the database. `config` can either be a subclass of
@@ -352,7 +343,7 @@ model with the same key is already in the database, or an
 `InvalidModelConfigException` if a dict was passed and Pydantic
 experienced a parse or validation error.
 
-### update_model(key, config) -> AnyModelConfig:
+### update_model(key, config) -> AnyModelConfig
 
 Given a key and a configuration, this will update the model
 configuration record in the database. `config` can be either a
@@ -370,33 +361,30 @@ The `ModelInstallService` class implements the
 shop for all your model install needs. It provides the following
 functionality:
 
-- Registering a model config record for a model already located on the
+* Registering a model config record for a model already located on the
   local filesystem, without moving it or changing its path.
   
-- Installing a model alreadiy located on the local filesystem, by
+* Installing a model alreadiy located on the local filesystem, by
   moving it into the InvokeAI root directory under the
   `models` folder (or wherever config parameter `models_dir`
   specifies).
 
-- Probing of models to determine their type, base type and other key
+* Probing of models to determine their type, base type and other key
   information.
   
-- Interface with the InvokeAI event bus to provide status updates on
+* Interface with the InvokeAI event bus to provide status updates on
   the download, installation and registration process.
   
-- Downloading a model from an arbitrary URL and installing it in
+* Downloading a model from an arbitrary URL and installing it in
   `models_dir`.
 
-- Special handling for Civitai model URLs which allow the user to
-  paste in a model page's URL or download link
-  
-- Special handling for HuggingFace repo_ids to recursively download
+* Special handling for HuggingFace repo_ids to recursively download
   the contents of the repository, paying attention to alternative
   variants such as fp16.
   
-- Saving tags and other metadata about the model into the invokeai database
+* Saving tags and other metadata about the model into the invokeai database
   when fetching from a repo that provides that type of information,
-  (currently only Civitai and HuggingFace).
+  (currently only HuggingFace).
   
 ### Initializing the installer
 
@@ -440,10 +428,8 @@ required parameters:
 | `app_config`         | InvokeAIAppConfig       | InvokeAI app configuration object |
 | `record_store`   | ModelRecordServiceBase  | Config record storage database |
 | `download_queue`   | DownloadQueueServiceBase  | Download queue object |
-| `metadata_store`   | Optional[ModelMetadataStore]  | Metadata storage object |
 |`session`           | Optional[requests.Session]    | Swap in a different Session object (usually for debugging) |
 
-
 Once initialized, the installer will provide the following methods:
 
 #### install_job = installer.heuristic_import(source, [config], [access_token])
@@ -457,12 +443,12 @@ The `source` is a string that can be any of these forms
 1. A path on the local filesystem (`C:\\users\\fred\\model.safetensors`)
 2. A Url pointing to a single downloadable model file (`https://civitai.com/models/58390/detail-tweaker-lora-lora`)
 3. A HuggingFace repo_id with any of the following formats:
-   - `model/name` -- entire model
-   - `model/name:fp32` -- entire model, using the fp32 variant
-   - `model/name:fp16:vae` -- vae submodel, using the fp16 variant
-   - `model/name::vae` -- vae submodel, using default precision
-   - `model/name:fp16:path/to/model.safetensors` -- an individual model file, fp16 variant
-   - `model/name::path/to/model.safetensors` -- an individual model file, default variant
+   * `model/name` -- entire model
+   * `model/name:fp32` -- entire model, using the fp32 variant
+   * `model/name:fp16:vae` -- vae submodel, using the fp16 variant
+   * `model/name::vae` -- vae submodel, using default precision
+   * `model/name:fp16:path/to/model.safetensors` -- an individual model file, fp16 variant
+   * `model/name::path/to/model.safetensors` -- an individual model file, default variant
 
 Note that by specifying a relative path to the top of the HuggingFace
 repo, you can download and install arbitrary models files.
@@ -566,7 +552,6 @@ details.
 This is used for a model that is located on a locally-accessible Posix
 filesystem, such as a local disk or networked fileshare.
 
-
 | **Argument**     | **Type**                     | **Default** | **Description**                           |
 |------------------|------------------------------|-------------|-------------------------------------------|
 | `path`           | str | Path                   | None        | Path to the model file or directory |
@@ -586,33 +571,7 @@ The `AnyHttpUrl` class can be imported from `pydantic.networks`.
 
 Ordinarily, no metadata is retrieved from these sources. However,
 there is special-case code in the installer that looks for HuggingFace
-and Civitai URLs and fetches the corresponding model metadata from
-the corresponding repo.
-
-#### CivitaiModelSource
-
-This is used for a model that is hosted by the Civitai web site.
-
-| **Argument**     | **Type**                     | **Default** | **Description**                           |
-|------------------|------------------------------|-------------|-------------------------------------------|
-| `version_id`     | int                          | None        | The ID of the particular version of the desired model. |
-| `access_token`   | str                          | None        | An access token needed to gain access to a subscriber's-only model. |
-
-Civitai has two model IDs, both of which are integers. The `model_id`
-corresponds to a collection of model versions that may different in
-arbitrary ways, such as derivation from different checkpoint training
-steps, SFW vs NSFW generation, pruned vs non-pruned, etc. The
-`version_id` points to a specific version. Please use the latter.
-
-Some Civitai models require an access token to download. These can be
-generated from the Civitai profile page of a logged-in
-account. Somewhat annoyingly, if you fail to provide the access token
-when downloading a model that needs it, Civitai generates a redirect
-to a login page rather than a 403 Forbidden error. The installer
-attempts to catch this event and issue an informative error
-message. Otherwise you will get an "unrecognized model suffix" error
-when the model prober tries to identify the type of the HTML login
-page.
+and fetches the corresponding model metadata from the corresponding repo.
 
 #### HFModelSource
 
@@ -625,7 +584,6 @@ HuggingFace has the most complicated `ModelSource` structure:
 | `subfolder`      | Path                         | None        | Look for the model in a subfolder of the repo. |
 | `access_token`   | str                          | None        | An access token needed to gain access to a subscriber's-only model. |
 
-
 The `repo_id` is the repository ID, such as `stabilityai/sdxl-turbo`.
 
 The `variant` is one of the various diffusers formats that HuggingFace
@@ -661,7 +619,6 @@ in. To download these files, you must provide an
 `HfFolder.get_token()` will be called to fill it in with the cached
 one.
 
-
 #### Monitoring the install job process
 
 When you create an install job with `import_model()`, it launches the
@@ -682,7 +639,6 @@ The `ModelInstallJob` class has the following structure:
 | `error_type`   | `str`           | Name of the exception that led to an error status |
 | `error`        | `str`           | Traceback of the error |
 
-
 If the `event_bus` argument was provided, events will also be
 broadcast to the InvokeAI event bus. The events will appear on the bus
 as an event of type `EventServiceBase.model_event`, a timestamp and
@@ -702,14 +658,13 @@ following keys:
 | `total_bytes`  | int       | Total size of all the files that make up the model |
 | `parts`        | List[Dict]| Information on the progress of the individual files that make up the model |
 
-
 The parts is a list of dictionaries that give information on each of
 the components pieces of the download. The dictionary's keys are
 `source`, `local_path`, `bytes` and `total_bytes`, and correspond to
 the like-named keys in the main event.
 
 Note that downloading events will not be issued for local models, and
-that downloading events occur *before* the running event.
+that downloading events occur _before_ the running event.
 
 ##### `model_install_running`
 
@@ -752,7 +707,6 @@ properties: `waiting`, `downloading`, `running`, `complete`, `errored`
 and `cancelled`, as well as `in_terminal_state`. The last will return
 True if the job is in the complete, errored or cancelled states.
 
-
 #### Model configuration and probing
 
 The install service uses the `invokeai.backend.model_manager.probe`
@@ -862,7 +816,6 @@ This method is similar to `unregister()`, but also unconditionally
 deletes the corresponding model weights file(s), regardless of whether
 they are inside or outside the InvokeAI models hierarchy.
 
-
 #### path = installer.download_and_cache(remote_source, [access_token], [timeout])
 
 This utility routine will download the model file located at source,
@@ -974,7 +927,7 @@ is in its lifecycle. Values are defined in the string enum
 `DownloadJobStatus`, a symbol available from
 `invokeai.app.services.download_manager`. Possible values are:
 
-| **Value**    |   **String Value**  | ** Description ** |
+| **Value**    |   **String Value**  | **Description** |
 |--------------|---------------------|-------------------|
 | `IDLE`         | idle              | Job created, but not submitted to the queue |
 | `ENQUEUED`     | enqueued          | Job is patiently waiting on the queue       |
@@ -1040,11 +993,11 @@ While a job is being downloaded, the queue will emit events at
 periodic intervals. A typical series of events during a successful
 download session will look like this:
 
-- enqueued
-- running
-- running
-- running
-- completed
+* enqueued
+* running
+* running
+* running
+* completed
 
 There will be a single enqueued event, followed by one or more running
 events, and finally one `completed`, `error` or `cancelled`
@@ -1053,12 +1006,12 @@ events.
 It is possible for a caller to pause download temporarily, in which
 case the events may look something like this:
 
-- enqueued
-- running
-- running
-- paused
-- running
-- completed
+* enqueued
+* running
+* running
+* paused
+* running
+* completed
 
 The download queue logs when downloads start and end (unless `quiet`
 is set to True at initialization time) but doesn't log any progress
@@ -1187,7 +1140,6 @@ and is equivalent to manually specifying a destination of
 Here is the full list of arguments that can be provided to
 `create_download_job()`:
 
-
 | **Argument**     | **Type**                     | **Default** | **Description**                           |
 |------------------|------------------------------|-------------|-------------------------------------------|
 | `source`         | Union[str, Path, AnyHttpUrl] |             | Download remote or local source           |
@@ -1266,51 +1218,30 @@ queue and have not yet reached a terminal state.
 
 The modules found under `invokeai.backend.model_manager.metadata`
 provide a straightforward API for fetching model metadatda from online
-repositories. Currently two repositories are supported: HuggingFace
-and Civitai. However, the modules are easily extended for additional
-repos, provided that they have defined APIs for metadata access.
+repositories. Currently only HuggingFace is supported. However, the
+modules are easily extended for additional repos, provided that they
+have defined APIs for metadata access.
 
 Metadata comprises any descriptive information that is not essential
 for getting the model to run. For example "author" is metadata, while
 "type", "base" and "format" are not. The latter fields are part of the
 model's config, as defined in `invokeai.backend.model_manager.config`.
 
-### Example Usage:
+### Example Usage
 
 ```
 from invokeai.backend.model_manager.metadata import (
    AnyModelRepoMetadata,
-   CivitaiMetadataFetch,
-   CivitaiMetadata
-   ModelMetadataStore,
 )
 # to access the initialized sql database
 from invokeai.app.api.dependencies import ApiDependencies
 
-civitai = CivitaiMetadataFetch()
+hf = HuggingFaceMetadataFetch()
 
 # fetch the metadata
-model_metadata = civitai.from_url("https://civitai.com/models/215796")
+model_metadata = hf.from_id("<repo_id>")
 
-# get some common metadata fields
-author = model_metadata.author
-tags = model_metadata.tags
-
-# get some Civitai-specific fields
-assert isinstance(model_metadata, CivitaiMetadata)
-
-trained_words = model_metadata.trained_words
-base_model = model_metadata.base_model_trained_on
-thumbnail = model_metadata.thumbnail_url
-
-# cache the metadata to the database using the key corresponding to
-# an existing model config record in the `model_config` table
-sql_cache = ModelMetadataStore(ApiDependencies.invoker.services.db)
-sql_cache.add_metadata('fb237ace520b6716adc98bcb16e8462c', model_metadata)
-
-# now we can search the database by tag, author or model name
-# matches will contain a list of model keys that match the search
-matches = sql_cache.search_by_tag({"tool", "turbo"})
+assert isinstance(model_metadata, HuggingFaceMetadata)
 ```
 
 ### Structure of the Metadata objects
@@ -1328,7 +1259,6 @@ This is the common base class for metadata:
 | `author`      | str           | Model's author |
 | `tags`        | Set[str]      | Model tags |
 
-
 Note that the model config record also has a `name` field. It is
 intended that the config record version be locally customizable, while
 the metadata version is read-only. However, enforcing this is expected
@@ -1348,53 +1278,14 @@ This descends from `ModelMetadataBase` and adds the following fields:
 | `last_modified`| datetime         | Date of last commit of this model to the repo |
 | `files`        | List[Path]       | List of the files in the model repo |
 
-
-#### `CivitaiMetadata`
-
-This descends from `ModelMetadataBase` and adds the following fields:
-
-| **Field Name** | **Type**        |  **Description** |
-|----------------|-----------------|------------------|
-| `type`         | Literal["civitai"]   | Used for the discriminated union of metadata classes|
-| `id`           | int                  | Civitai model id |
-| `version_name` | str                  | Name of this version of the model (distinct from model name) |
-| `version_id`   | int                  | Civitai model version id (distinct from model id) |
-| `created`      | datetime             | Date this version of the model was created |
-| `updated`      | datetime             | Date this version of the model was last updated |
-| `published`    | datetime             | Date this version of the model was published to Civitai |
-| `description`  | str                  | Model description. Quite verbose and contains HTML tags |
-| `version_description` | str           | Model version description, usually describes changes to the model |
-| `nsfw`         | bool                 | Whether the model tends to generate NSFW content |
-| `restrictions` | LicenseRestrictions  | An object that describes what is and isn't allowed with this model |
-| `trained_words`| Set[str]             | Trigger words for this model, if any |
-| `download_url` | AnyHttpUrl           | URL for downloading this version of the model |
-| `base_model_trained_on` | str         | Name of the model that this version was trained on |
-| `thumbnail_url` | AnyHttpUrl          | URL to access a representative thumbnail image of the model's output |
-| `weight_min`    | int                 | For LoRA sliders, the minimum suggested weight to apply |
-| `weight_max`    | int                 | For LoRA sliders, the maximum suggested weight to apply |
-
-Note that `weight_min` and `weight_max` are not currently populated
-and take the default values of (-1.0, +2.0). The issue is that these
-values aren't part of the structured data but appear in the text
-description. Some regular expression or LLM coding may be able to
-extract these values.
-
-Also be aware that `base_model_trained_on` is free text and doesn't
-correspond to our `ModelType` enum.
-
-`CivitaiMetadata` also defines some convenience properties relating to
-licensing restrictions: `credit_required`, `allow_commercial_use`,
-`allow_derivatives` and `allow_different_license`.
-
 #### `AnyModelRepoMetadata`
 
-This is a discriminated Union of `CivitaiMetadata` and
-`HuggingFaceMetadata`.
+This is a discriminated Union of `HuggingFaceMetadata`.
 
 ### Fetching Metadata from Online Repos
 
-The `HuggingFaceMetadataFetch` and `CivitaiMetadataFetch` classes will
-retrieve metadata from their corresponding repositories and return
+The `HuggingFaceMetadataFetch` class will
+retrieve metadata from its corresponding repository and return
 `AnyModelRepoMetadata` objects. Their base class
 `ModelMetadataFetchBase` is an abstract class that defines two
 methods: `from_url()` and `from_id()`. The former accepts the type of
@@ -1412,98 +1303,17 @@ provide a `requests.Session` argument. This allows you to customize
 the low-level HTTP fetch requests and is used, for instance, in the
 testing suite to avoid hitting the internet.
 
-The HuggingFace and Civitai fetcher subclasses add additional
-repo-specific fetching methods:
-
+The HuggingFace fetcher subclass add additional repo-specific fetching methods:
 
 #### HuggingFaceMetadataFetch
 
 This overrides its base class `from_json()` method to return a
 `HuggingFaceMetadata` object directly.
 
-#### CivitaiMetadataFetch
-
-This adds the following methods:
-
-`from_civitai_modelid()` This takes the ID of a model, finds the
-default version of the model, and then retrieves the metadata for
-that version, returning a `CivitaiMetadata` object directly.
-
-`from_civitai_versionid()` This takes the ID of a model version and
-retrieves its metadata. Functionally equivalent to `from_id()`, the
-only difference is that it returna a `CivitaiMetadata` object rather
-than an `AnyModelRepoMetadata`.
-
-
 ### Metadata Storage
 
-The `ModelMetadataStore` provides a simple facility to store model
-metadata in the `invokeai.db` database. The data is stored as a JSON
-blob, with a few common fields (`name`, `author`, `tags`) broken out
-to be searchable. 
-
-When a metadata object is saved to the database, it is identified
-using the model key, _and this key must correspond to an existing
-model key in the model_config table_. There is a foreign key integrity
-constraint between the `model_config.id` field and the
-`model_metadata.id` field such that if you attempt to save metadata
-under an unknown key, the attempt will result in an
-`UnknownModelException`. Likewise, when a model is deleted from
-`model_config`, the deletion of the corresponding metadata record will
-be triggered.
-
-Tags are stored in a normalized fashion in the tables `model_tags` and
-`tags`. Triggers keep the tag table in sync with the `model_metadata`
-table.
-
-To create the storage object, initialize it with the InvokeAI
-`SqliteDatabase` object. This is often done this way:
-
-```
-from invokeai.app.api.dependencies import ApiDependencies
-metadata_store = ModelMetadataStore(ApiDependencies.invoker.services.db)
-```
-
-You can then access the storage with the following methods:
-
-#### `add_metadata(key, metadata)`
-
-Add the metadata using a previously-defined model key.
-
-There is currently no `delete_metadata()` method. The metadata will
-persist until the matching config is deleted from the `model_config`
-table.
-
-#### `get_metadata(key) -> AnyModelRepoMetadata`
-
-Retrieve the metadata corresponding to the model key.
-
-#### `update_metadata(key, new_metadata)`
-
-Update an existing metadata record with new metadata.
-
-#### `search_by_tag(tags: Set[str]) -> Set[str]`
-
-Given a set of tags, find models that are tagged with them. If
-multiple tags are provided then a matching model must be tagged with
-*all* the tags in the set. This method returns a set of model keys and
-is intended to be used in conjunction with the `ModelRecordService`:
-
-```
-model_config_store = ApiDependencies.invoker.services.model_records
-matches = metadata_store.search_by_tag({'license:other'})
-models = [model_config_store.get(x) for x in matches]
-```
-
-#### `search_by_name(name: str) -> Set[str]
-
-Find all model metadata records that have the given name and return a
-set of keys to the corresponding model config objects.
-
-#### `search_by_author(author: str) -> Set[str]
-
-Find all model metadata records that have the given author and return
-a set of keys to the corresponding model config objects.
+The `ModelConfigBase` stores this response in the `source_api_response` field
+as a JSON blob.
 
 ***
 
@@ -1567,7 +1377,6 @@ The returned `LoadedModel` object contains a copy of the configuration
 record returned by the model record `get_model()` method, as well as
 the in-memory loaded model:
 
-
 | **Attribute Name** | **Type**        |  **Description** |
 |----------------|-----------------|------------------|
 | `config`       | AnyModelConfig         | A copy of the model's configuration record for retrieving base type, etc. |
@@ -1581,7 +1390,6 @@ return `AnyModel`, a Union `ModelMixin`, `torch.nn.Module`,
 models, `EmbeddingModelRaw` is used for LoRA and TextualInversion
 models. The others are obvious.
 
-
 `LoadedModel` acts as a context manager. The context loads the model
 into the execution device (e.g. VRAM on CUDA systems), locks the model
 in the execution device for the duration of the context, and returns
@@ -1595,9 +1403,9 @@ with model_info as vae:
 
 `get_model_by_key()` may raise any of the following exceptions:
 
-- `UnknownModelException`   -- key not in database
-- `ModelNotFoundException`  -- key in database but model not found at path
-- `NotImplementedException` -- the loader doesn't know how to load this type of model
+* `UnknownModelException`   -- key not in database
+* `ModelNotFoundException`  -- key in database but model not found at path
+* `NotImplementedException` -- the loader doesn't know how to load this type of model
   
 ### Emitting model loading events
 
@@ -1724,6 +1532,7 @@ object, or in `context.services.model_manager` from within an
 invocation.
 
 In the examples below, we have retrieved the manager using:
+
 ```
 mm = ApiDependencies.invoker.services.model_manager
 ```

docs/contributing/frontend/OVERVIEW.md

@@ -0,0 +1,133 @@
+# Invoke UI
+
+Invoke's UI is made possible by many contributors and open-source libraries. Thank you!
+
+## Dev environment
+
+### Setup
+
+1. Install [node] and [pnpm].
+1. Run `pnpm i` to install all packages.
+
+#### Run in dev mode
+
+1. From `invokeai/frontend/web/`, run `pnpm dev`.
+1. From repo root, run `python scripts/invokeai-web.py`.
+1. Point your browser to the dev server address, e.g. <http://localhost:5173/>
+
+### Package scripts
+
+- `dev`: run the frontend in dev mode, enabling hot reloading
+- `build`: run all checks (madge, eslint, prettier, tsc) and then build the frontend
+- `typegen`: generate types from the OpenAPI schema (see [Type generation])
+- `lint:dpdm`: check circular dependencies
+- `lint:eslint`: check code quality
+- `lint:prettier`: check code formatting
+- `lint:tsc`: check type issues
+- `lint:knip`: check for unused exports or objects (failures here are just suggestions, not hard fails)
+- `lint`: run all checks concurrently
+- `fix`: run `eslint` and `prettier`, fixing fixable issues
+
+### Type generation
+
+We use [openapi-typescript] to generate types from the app's OpenAPI schema.
+
+The generated types are committed to the repo in [schema.ts].
+
+```sh
+# from the repo root, start the server
+python scripts/invokeai-web.py
+# from invokeai/frontend/web/, run the script
+pnpm typegen
+```
+
+### Localization
+
+We use [i18next] for localization, but translation to languages other than English happens on our [Weblate] project.
+
+Only the English source strings should be changed on this repo.
+
+### VSCode
+
+#### Example debugger config
+
+```jsonc
+{
+  "version": "0.2.0",
+  "configurations": [
+    {
+      "type": "chrome",
+      "request": "launch",
+      "name": "Invoke UI",
+      "url": "http://localhost:5173",
+      "webRoot": "${workspaceFolder}/invokeai/frontend/web"
+    }
+  ]
+}
+```
+
+#### Remote dev
+
+We've noticed an intermittent timeout issue with the VSCode remote dev port forwarding.
+
+We suggest disabling the editor's port forwarding feature and doing it manually via SSH:
+
+```sh
+ssh -L 9090:localhost:9090 -L 5173:localhost:5173 user@host
+```
+
+## Contributing Guidelines
+
+Thanks for your interest in contributing to the Invoke Web UI!
+
+Please follow these guidelines when contributing.
+
+### Check in before investing your time
+
+Please check in before you invest your time on anything besides a trivial fix, in case it conflicts with ongoing work or isn't aligned with the vision for the app.
+
+If a feature request or issue doesn't already exist for the thing you want to work on, please create one.
+
+Ping `@psychedelicious` on [discord] in the `#frontend-dev` channel or in the feature request / issue you want to work on - we're happy to chat.
+
+### Code conventions
+
+- This is a fairly complex app with a deep component tree. Please use memoization (`useCallback`, `useMemo`, `memo`) with enthusiasm.
+- If you need to add some global, ephemeral state, please use [nanostores] if possible.
+- Be careful with your redux selectors. If they need to be parameterized, consider creating them inside a `useMemo`.
+- Feel free to use `lodash` (via `lodash-es`) to make the intent of your code clear.
+- Please add comments describing the "why", not the "how" (unless it is really arcane).
+
+### Commit format
+
+Please use the [conventional commits] spec for the web UI, with a scope of "ui":
+
+- `chore(ui): bump deps`
+- `chore(ui): lint`
+- `feat(ui): add some cool new feature`
+- `fix(ui): fix some bug`
+
+### Submitting a PR
+
+- Ensure your branch is tidy. Use an interactive rebase to clean up the commit history and reword the commit messages if they are not descriptive.
+- Run `pnpm lint`. Some issues are auto-fixable with `pnpm fix`.
+- Fill out the PR form when creating the PR.
+  - It doesn't need to be super detailed, but a screenshot or video is nice if you changed something visually.
+  - If a section isn't relevant, delete it. There are no UI tests at this time.
+
+## Other docs
+
+- [Workflows - Design and Implementation]
+- [State Management]
+
+[node]: https://nodejs.org/en/download/
+[pnpm]: https://github.com/pnpm/pnpm
+[discord]: https://discord.gg/ZmtBAhwWhy
+[i18next]: https://github.com/i18next/react-i18next
+[Weblate]: https://hosted.weblate.org/engage/invokeai/
+[openapi-typescript]: https://github.com/drwpow/openapi-typescript
+[Type generation]: #type-generation
+[schema.ts]: https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/frontend/web/src/services/api/schema.ts
+[conventional commits]: https://www.conventionalcommits.org/en/v1.0.0/
+[Workflows - Design and Implementation]: ./WORKFLOWS.md
+[State Management]: ./STATE_MGMT.md

docs/contributing/frontend/WORKFLOWS.md

@@ -1,40 +1,5 @@
 # Workflows - Design and Implementation
 
-<!-- @import "[TOC]" {cmd="toc" depthFrom=1 depthTo=6 orderedList=false} -->
-
-<!-- code_chunk_output -->
-
-- [Workflows - Design and Implementation](#workflows---design-and-implementation)
-  - [Design](#design)
-    - [Linear UI](#linear-ui)
-    - [Workflow Editor](#workflow-editor)
-      - [Workflows](#workflows)
-        - [Workflow -> reactflow state -> InvokeAI graph](#workflow---reactflow-state---invokeai-graph)
-        - [Nodes vs Invocations](#nodes-vs-invocations)
-        - [Workflow Linear View](#workflow-linear-view)
-      - [OpenAPI Schema](#openapi-schema)
-        - [Field Instances and Templates](#field-instances-and-templates)
-        - [Stateful vs Stateless Fields](#stateful-vs-stateless-fields)
-        - [Collection and Polymorphic Fields](#collection-and-polymorphic-fields)
-  - [Implementation](#implementation)
-    - [zod Schemas and Types](#zod-schemas-and-types)
-    - [OpenAPI Schema Parsing](#openapi-schema-parsing)
-      - [Parsing Field Types](#parsing-field-types)
-        - [Primitive Types](#primitive-types)
-        - [Complex Types](#complex-types)
-        - [Collection Types](#collection-types)
-        - [Collection or Scalar Types](#collection-or-scalar-types)
-        - [Optional Fields](#optional-fields)
-      - [Building Field Input Templates](#building-field-input-templates)
-      - [Building Field Output Templates](#building-field-output-templates)
-    - [Managing reactflow State](#managing-reactflow-state)
-      - [Building Nodes and Edges](#building-nodes-and-edges)
-      - [Building a Workflow](#building-a-workflow)
-      - [Loading a Workflow](#loading-a-workflow)
-    - [Workflow Migrations](#workflow-migrations)
-
-<!-- /code_chunk_output -->
-
 > This document describes, at a high level, the design and implementation of workflows in the InvokeAI frontend. There are a substantial number of implementation details not included, but which are hopefully clear from the code.
 
 InvokeAI's backend uses graphs, composed of **nodes** and **edges**, to process data and generate images.
@@ -152,13 +117,13 @@ Stateless fields do not store their value in the node, so their field instances
 
 "Custom" fields will always be treated as stateless fields.
 
-##### Collection and Polymorphic Fields
+##### Collection and Scalar Fields
 
-Field types have a name and two flags which may identify it as a **collection** or **polymorphic** field.
+Field types have a name and two flags which may identify it as a **collection** or **collection or scalar** field.
 
-If a field is annotated in python as a list, its field type is parsed and flagged as a collection type (e.g. `list[int]`).
+If a field is annotated in python as a list, its field type is parsed and flagged as a **collection** type (e.g. `list[int]`).
 
-If it is annotated as a union of a type and list, the type will be flagged as a polymorphic type (e.g. `Union[int, list[int]]`). Fields may not be unions of different types (e.g. `Union[int, list[str]]` and `Union[int, str]` are not allowed).
+If it is annotated as a union of a type and list, the type will be flagged as a **collection or scalar** type (e.g. `Union[int, list[int]]`). Fields may not be unions of different types (e.g. `Union[int, list[str]]` and `Union[int, str]` are not allowed).
 
 ## Implementation
 
@@ -338,13 +303,13 @@ Migration logic is in [migrations.ts].
 [reactflow]: https://github.com/xyflow/xyflow 'reactflow'
 [reactflow-concepts]: https://reactflow.dev/learn/concepts/terms-and-definitions
 [reactflow-events]: https://reactflow.dev/api-reference/react-flow#event-handlers
-[buildWorkflow.ts]: ../src/features/nodes/util/workflow/buildWorkflow.ts
-[nodesSlice.ts]: ../src/features/nodes/store/nodesSlice.ts
-[buildLinearTextToImageGraph.ts]: ../src/features/nodes/util/graph/buildLinearTextToImageGraph.ts
-[buildNodesGraph.ts]: ../src/features/nodes/util/graph/buildNodesGraph.ts
-[buildInvocationNode.ts]: ../src/features/nodes/util/node/buildInvocationNode.ts
-[validateWorkflow.ts]: ../src/features/nodes/util/workflow/validateWorkflow.ts
-[migrations.ts]: ../src/features/nodes/util/workflow/migrations.ts
-[parseSchema.ts]: ../src/features/nodes/util/schema/parseSchema.ts
-[buildFieldInputTemplate.ts]: ../src/features/nodes/util/schema/buildFieldInputTemplate.ts
-[buildFieldOutputTemplate.ts]: ../src/features/nodes/util/schema/buildFieldOutputTemplate.ts
+[buildWorkflow.ts]: https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/frontend/web/src/features/nodes/util/workflow/buildWorkflow.ts
+[nodesSlice.ts]: https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/frontend/web/src/features/nodes/store/nodesSlice.ts
+[buildLinearTextToImageGraph.ts]: https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/frontend/web/src/features/nodes/util/graph/buildLinearTextToImageGraph.ts
+[buildNodesGraph.ts]: https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/frontend/web/src/features/nodes/util/graph/buildNodesGraph.ts
+[buildInvocationNode.ts]: https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/frontend/web/src/features/nodes/util/node/buildInvocationNode.ts
+[validateWorkflow.ts]: https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/frontend/web/src/features/nodes/util/workflow/validateWorkflow.ts
+[migrations.ts]: https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/frontend/web/src/features/nodes/util/workflow/migrations.ts
+[parseSchema.ts]: https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/frontend/web/src/features/nodes/util/schema/parseSchema.ts
+[buildFieldInputTemplate.ts]: https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/frontend/web/src/features/nodes/util/schema/buildFieldInputTemplate.ts
+[buildFieldOutputTemplate.ts]: https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/frontend/web/src/features/nodes/util/schema/buildFieldOutputTemplate.ts

docs/features/CONFIGURATION.md

@@ -31,18 +31,18 @@ be referred to as ROOT.
 To find its root directory, InvokeAI uses the following recipe:
 
 1. It first looks for the argument `--root <path>` on the command line
-it was launched from, and uses the indicated path if present.
+   it was launched from, and uses the indicated path if present.
 
 2. Next it looks for the environment variable INVOKEAI_ROOT, and uses
-the directory path found there if present.
+   the directory path found there if present.
 
 3. If neither of these are present, then InvokeAI looks for the
-folder containing the `.venv` Python virtual environment directory for
-the currently active environment. This directory is checked for files
-expected inside the InvokeAI root before it is used.
+   folder containing the `.venv` Python virtual environment directory for
+   the currently active environment. This directory is checked for files
+   expected inside the InvokeAI root before it is used.
 
 4. Finally, InvokeAI looks for a directory in the current user's home
-directory named `invokeai`.
+   directory named `invokeai`.
 
 #### Reading the InvokeAI Configuration File
 
@@ -149,104 +149,75 @@ usage: InvokeAI [-h] [--host HOST] [--port PORT] [--allow_origins [ALLOW_ORIGINS
 
 ## The Configuration Settings
 
-The configuration settings are divided into several distinct
-groups in `invokeia.yaml`:
+The config is managed by the `InvokeAIAppConfig` class, which is a pydantic model. The below docs are autogenerated from the class.
 
-### Web Server
+When editing your `invokeai.yaml` file, you'll need to put settings under their appropriate group. The group for each setting is denoted in the table below.
 
-| Setting             | Default Value | Description                                                                                                                |
-|---------------------|---------------|----------------------------------------------------------------------------------------------------------------------------|
-| `host`              | `localhost`   | Name or IP address of the network interface that the web server will listen on                                             |
-| `port`              | `9090`        | Network port number that the web server will listen on                                                                     |
-| `allow_origins`     | `[]`          | A list of host names or IP addresses that are allowed to connect to the InvokeAI API in the format `['host1','host2',...]` |
-| `allow_credentials` | `true`        | Require credentials for a foreign host to access the InvokeAI API (don't change this)                                      |
-| `allow_methods`     | `*`           | List of HTTP methods ("GET", "POST") that the web server is allowed to use when accessing the API                          |
-| `allow_headers`     | `*`           | List of HTTP headers that the web server will accept when accessing the API                                                |
-| `ssl_certfile`      | null          | Path to an SSL certificate file, used to enable HTTPS.                                                                     |
-| `ssl_keyfile`       | null          | Path to an SSL keyfile, if the key is not included in the certificate file.                                                |
+Following the table are additional explanations for certain settings.
 
-The documentation for InvokeAI's API can be accessed by browsing to the following URL: [http://localhost:9090/docs].
+<!-- prettier-ignore-start -->
+::: invokeai.app.services.config.config_default.InvokeAIAppConfig
+    options:
+        heading_level: 3
+        members: false
+<!-- prettier-ignore-end -->
 
-### Features
+### Model Marketplace API Keys
 
-These configuration settings allow you to enable and disable various InvokeAI features:
+Some model marketplaces require an API key to download models. You can provide a URL pattern and appropriate token in your `invokeai.yaml` file to provide that API key.
 
-| Setting  | Default Value  |  Description |
-|----------|----------------|--------------|
-| `esrgan`     | `true`      | Activate the ESRGAN upscaling options|
-| `internet_available` | `true`     | When a resource is not available locally, try to fetch it via the internet |
-| `log_tokenization` | `false`      | Before each text2image generation, print a color-coded representation of the prompt to the console; this can help understand why a prompt is not working as expected |
-| `patchmatch` | `true`     | Activate the "patchmatch" algorithm for improved inpainting |
+The pattern can be any valid regex (you may need to surround the pattern with quotes):
 
-### Generation
+```yaml
+InvokeAI:
+  Model Install:
+    remote_api_tokens:
+      # Any URL containing `models.com` will automatically use `your_models_com_token`
+      - url_regex: models.com
+        token: your_models_com_token
+      # Any URL matching this contrived regex will use `some_other_token`
+      - url_regex: '^[a-z]{3}whatever.*\.com$'
+        token: some_other_token
+```
 
-These options tune InvokeAI's memory and performance characteristics.
+The provided token will be added as a `Bearer` token to the network requests to download the model files. As far as we know, this works for all model marketplaces that require authorization.
 
-| Setting               | Default Value | Description                                                                                                                                                                                                                                                      |
-|-----------------------|---------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
-| `sequential_guidance` | `false`       | Calculate guidance in serial rather than in parallel, lowering memory requirements at the cost of some performance loss                                                                                                                                          |
-| `attention_type`      | `auto`        | Select the type of attention to use. One of `auto`,`normal`,`xformers`,`sliced`, or `torch-sdp`                                                                                                                                                                  |
-| `attention_slice_size` | `auto`       | When "sliced" attention is selected, set the slice size. One of `auto`, `balanced`, `max` or the integers 1-8|
-| `force_tiled_decode`  | `false`       | Force the VAE step to decode in tiles, reducing memory consumption at the cost of performance |
+### Model Hashing
 
-### Device
+Models are hashed during installation, providing a stable identifier for models across all platforms. The default algorithm is `blake3`, with a multi-threaded implementation.
 
-These options configure the generation execution device.
+If your models are stored on a spinning hard drive, we suggest using `blake3_single`, the single-threaded implementation. The hashes are the same, but it's much faster on spinning disks.
 
-| Setting               | Default Value | Description                                                                                                                                                                                                                                                      |
-|-----------------------|---------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
-| `device`           | `auto`        | Preferred execution device. One of `auto`, `cpu`, `cuda`, `cuda:1`, `mps`. `auto` will choose the device depending on the hardware platform and the installed torch capabilities. |
-| `precision`           | `auto`        | Floating point precision. One of `auto`, `float16` or `float32`. `float16` will consume half the memory of `float32` but produce slightly lower-quality images. The `auto` setting will guess the proper precision based on your video card and operating system |
+```yaml
+InvokeAI:
+  Model Install:
+    hashing_algorithm: blake3_single
+```
 
+Model hashing is a one-time operation, but it may take a couple minutes to hash a large model collection. You may opt out of model hashing entirely by setting the algorithm to `random`.
+
+```yaml
+InvokeAI:
+  Model Install:
+    hashing_algorithm: random
+```
+
+Most common algorithms are supported, like `md5`, `sha256`, and `sha512`. These are typically much, much slower than `blake3`.
 
 ### Paths
 
 These options set the paths of various directories and files used by
-InvokeAI. Relative paths are interpreted relative to INVOKEAI_ROOT, so
-if INVOKEAI_ROOT is `/home/fred/invokeai` and the path is
+InvokeAI. Relative paths are interpreted relative to the root directory, so
+if root is `/home/fred/invokeai` and the path is
 `autoimport/main`, then the corresponding directory will be located at
 `/home/fred/invokeai/autoimport/main`.
 
-| Setting  | Default Value  |  Description |
-|----------|----------------|--------------|
-| `autoimport_dir` | `autoimport/main`     | At startup time, read and import any main model files found in this directory |
-| `lora_dir` | `autoimport/lora`     | At startup time, read and import any LoRA/LyCORIS models found in this directory |
-| `embedding_dir` | `autoimport/embedding`  | At startup time, read and import any textual inversion (embedding) models found in this directory |
-| `controlnet_dir` | `autoimport/controlnet`  | At startup time, read and import any ControlNet models found in this directory |
-| `conf_path` | `configs/models.yaml`  | Location of the `models.yaml` model configuration file |
-| `models_dir` | `models`  | Location of the directory containing models installed by InvokeAI's model manager |
-| `legacy_conf_dir` | `configs/stable-diffusion`  | Location of the directory containing the .yaml configuration files for legacy checkpoint models |
-| `db_dir` | `databases`  | Location of the directory containing InvokeAI's image, schema and session database |
-| `outdir` | `outputs`  | Location of the directory in which the gallery of generated and uploaded images will be stored |
-| `use_memory_db` | `false`  | Keep database information in memory rather than on disk; this will not preserve image gallery information across restarts |
-
-Note that the autoimport directories will be searched recursively,
+Note that the autoimport directory will be searched recursively,
 allowing you to organize the models into folders and subfolders in any
-way you wish. In addition, while we have split up autoimport
-directories by the type of model they contain, this isn't
-necessary. You can combine different model types in the same folder
-and InvokeAI will figure out what they are. So you can easily use just
-one autoimport directory by commenting out the unneeded paths:
-
-```
-Paths:
-  autoimport_dir: autoimport
-#  lora_dir: null
-#  embedding_dir: null
-#  controlnet_dir: null
-```
+way you wish.
 
 ### Logging
 
-These settings control the information, warning, and debugging
-messages printed to the console log while InvokeAI is running:
-
-| Setting  | Default Value  |  Description |
-|----------|----------------|--------------|
-| `log_handlers` | `console` | This controls where log messages are sent, and can be a list of one or more destinations. Values include `console`, `file`, `syslog` and `http`. These are described in more detail below |
-| `log_format` | `color` | This controls the formatting of the log messages. Values are `plain`, `color`, `legacy` and `syslog` |
-| `log_level`  | `debug` | This filters messages according to the level of severity and can be one of `debug`, `info`, `warning`, `error` and `critical`. For example, setting to `warning` will display all messages at the warning level or higher, but won't display "debug" or "info" messages |
-
 Several different log handler destinations are available, and multiple destinations are supported by providing a list:
 
 ```
@@ -256,9 +227,9 @@ Several different log handler destinations are available, and multiple destinati
      - file=/var/log/invokeai.log
 ```
 
-* `console` is the default. It prints log messages to the command-line window from which InvokeAI was launched.
+- `console` is the default. It prints log messages to the command-line window from which InvokeAI was launched.
 
-* `syslog` is only available on Linux and Macintosh systems. It uses
+- `syslog` is only available on Linux and Macintosh systems. It uses
   the operating system's "syslog" facility to write log file entries
   locally or to a remote logging machine. `syslog` offers a variety
   of configuration options:
@@ -271,7 +242,7 @@ Several different log handler destinations are available, and multiple destinati
                         - Log to LAN-connected server "fredserver" using the facility LOG_USER and datagram packets.
 ```
 
-* `http` can be used to log to a remote web server. The server must be
+- `http` can be used to log to a remote web server. The server must be
   properly configured to receive and act on log messages. The option
   accepts the URL to the web server, and a `method` argument
   indicating whether the message should be submitted using the GET or
@@ -283,7 +254,7 @@ Several different log handler destinations are available, and multiple destinati
 
 The `log_format` option provides several alternative formats:
 
-* `color`    - default format providing time, date and a message, using text colors to distinguish different log severities
-* `plain`    - same as above, but monochrome text only
-* `syslog`   - the log level and error message only, allowing the syslog system to attach the time and date
-* `legacy`   - a format similar to the one used by the legacy 2.3 InvokeAI releases.
+- `color` - default format providing time, date and a message, using text colors to distinguish different log severities
+- `plain` - same as above, but monochrome text only
+- `syslog` - the log level and error message only, allowing the syslog system to attach the time and date
+- `legacy` - a format similar to the one used by the legacy 2.3 InvokeAI releases.

docs/features/DATABASE.md

@@ -0,0 +1,35 @@
+---
+title: Database
+---
+
+# Invoke's SQLite Database
+
+Invoke uses a SQLite database to store image, workflow, model, and execution data.
+
+We take great care to ensure your data is safe, by utilizing transactions and a database migration system.
+
+Even so, when testing an prerelease version of the app, we strongly suggest either backing up your database or using an in-memory database. This ensures any prelease hiccups or databases schema changes will not cause problems for your data.
+
+## Database Backup
+
+Backing up your database is very simple. Invoke's data is stored in an `$INVOKEAI_ROOT` directory - where your `invoke.sh`/`invoke.bat` and `invokeai.yaml` files live.
+
+To back up your database, copy the `invokeai.db` file from `$INVOKEAI_ROOT/databases/invokeai.db` to somewhere safe.
+
+If anything comes up during prelease testing, you can simply copy your backup back into `$INVOKEAI_ROOT/databases/`.
+
+## In-Memory Database
+
+SQLite can run on an in-memory database. Your existing database is untouched when this mode is enabled, but your existing data won't be accessible.
+
+This is very useful for testing, as there is no chance of a database change modifying your "physical" database.
+
+To run Invoke with a memory database, edit your `invokeai.yaml` file, and add `use_memory_db: true` to the `Paths:` stanza:
+
+```yaml
+InvokeAI:
+  Development:
+    use_memory_db: true
+```
+
+Delete this line (or set it to `false`) to use your main database.

docs/nodes/INVOCATION_API.md

@@ -0,0 +1,63 @@
+# Invocation API
+
+Each invocation's `invoke` method is provided a single arg - the Invocation
+Context.
+
+This object provides access to various methods, used to interact with the
+application. Loading and saving images, logging messages, etc.
+
+!!! warning ""
+
+    This API may shift slightly until the release of v4.0.0 as we work through a few final updates to the Model Manager.
+
+```py
+class MyInvocation(BaseInvocation):
+  ...
+  def invoke(self, context: InvocationContext) -> ImageOutput:
+      image_pil = context.images.get_pil(image_name)
+      # Do something to the image
+      image_dto = context.images.save(image_pil)
+      # Log a message
+      context.logger.info(f"Did something cool, image saved!")
+      ...
+```
+
+The full API is documented below.
+
+## Invocation Mixins
+
+Two important mixins are provided to facilitate working with metadata and gallery boards.
+
+### `WithMetadata`
+
+Inherit from this class (in addition to `BaseInvocation`) to add a `metadata` input to your node. When you do this, you can access the metadata dict from `self.metadata` in the `invoke()` function.
+
+The dict will be populated via the node's input, and you can add any metadata you'd like to it. When you call `context.images.save()`, if the metadata dict has any data, it be automatically embedded in the image.
+
+### `WithBoard`
+
+Inherit from this class (in addition to `BaseInvocation`) to add a `board` input to your node. This renders as a drop-down to select a board. The user's selection will be accessible from `self.board` in the `invoke()` function.
+
+When you call `context.images.save()`, if a board was selected, the image will added to that board as it is saved.
+
+<!-- prettier-ignore-start -->
+::: invokeai.app.services.shared.invocation_context.InvocationContext
+    options:
+        members: false
+
+::: invokeai.app.services.shared.invocation_context.ImagesInterface
+
+::: invokeai.app.services.shared.invocation_context.TensorsInterface
+
+::: invokeai.app.services.shared.invocation_context.ConditioningInterface
+
+::: invokeai.app.services.shared.invocation_context.ModelsInterface
+
+::: invokeai.app.services.shared.invocation_context.LoggerInterface
+
+::: invokeai.app.services.shared.invocation_context.ConfigInterface
+
+::: invokeai.app.services.shared.invocation_context.UtilInterface
+
+::: invokeai.app.services.shared.invocation_context.BoardsInterface
+<!-- prettier-ignore-end -->

docs/nodes/NODES_MIGRATION_V3_V4.md

@@ -0,0 +1,148 @@
+# Invoke v4.0.0 Nodes API Migration guide
+
+Invoke v4.0.0 is versioned as such due to breaking changes to the API utilized
+by nodes, both core and custom.
+
+## Motivation
+
+Prior to v4.0.0, the `invokeai` python package has not be set up to be utilized
+as a library. That is to say, it didn't have any explicitly public API, and node
+authors had to work with the unstable internal application API.
+
+v4.0.0 introduces a stable public API for nodes.
+
+## Changes
+
+There are two node-author-facing changes:
+
+1. Import Paths
+1. Invocation Context API
+
+### Import Paths
+
+All public objects are now exported from `invokeai.invocation_api`:
+
+```py
+# Old
+from invokeai.app.invocations.baseinvocation import (
+    BaseInvocation,
+    InputField,
+    InvocationContext,
+    invocation,
+)
+from invokeai.app.invocations.primitives import ImageField
+
+# New
+from invokeai.invocation_api import (
+    BaseInvocation,
+    ImageField,
+    InputField,
+    InvocationContext,
+    invocation,
+)
+```
+
+It's possible that we've missed some classes you need in your node. Please let
+us know if that's the case.
+
+### Invocation Context API
+
+Most nodes utilize the Invocation Context, an object that is passed to the
+`invoke` that provides access to data and services a node may need.
+
+Until now, that object and the services it exposed were internal. Exposing them
+to nodes means that changes to our internal implementation could break nodes.
+The methods on the services are also often fairly complicated and allowed nodes
+to footgun.
+
+In v4.0.0, this object has been refactored to be much simpler.
+
+See [INVOCATION_API](./INVOCATION_API.md) for full details of the API.
+
+!!! warning ""
+
+    This API may shift slightly until the release of v4.0.0 as we work through a few final updates to the Model Manager.
+
+#### Improved Service Methods
+
+The biggest offender was the image save method:
+
+```py
+# Old
+image_dto = context.services.images.create(
+    image=image,
+    image_origin=ResourceOrigin.INTERNAL,
+    image_category=ImageCategory.GENERAL,
+    node_id=self.id,
+    session_id=context.graph_execution_state_id,
+    is_intermediate=self.is_intermediate,
+    metadata=self.metadata,
+    workflow=context.workflow,
+)
+
+# New
+image_dto = context.images.save(image=image)
+```
+
+Other methods are simplified, or enhanced with additional functionality:
+
+```py
+# Old
+image = context.services.images.get_pil_image(image_name)
+
+# New
+image = context.images.get_pil(image_name)
+image_cmyk = context.images.get_pil(image_name, "CMYK")
+```
+
+We also had some typing issues around tensors:
+
+```py
+# Old
+# `latents` typed as `torch.Tensor`, but could be `ConditioningFieldData`
+latents = context.services.latents.get(self.latents.latents_name)
+# `data` typed as `torch.Tenssor,` but could be `ConditioningFieldData`
+context.services.latents.save(latents_name, data)
+
+# New - separate methods for tensors and conditioning data w/ correct typing
+# Also, the service generates the names
+tensor_name = context.tensors.save(tensor)
+tensor = context.tensors.load(tensor_name)
+# For conditioning
+cond_name = context.conditioning.save(cond_data)
+cond_data = context.conditioning.load(cond_name)
+```
+
+#### Output Construction
+
+Core Outputs have builder functions right on them - no need to manually
+construct these objects, or use an extra utility:
+
+```py
+# Old
+image_output = ImageOutput(
+    image=ImageField(image_name=image_dto.image_name),
+    width=image_dto.width,
+    height=image_dto.height,
+)
+latents_output = build_latents_output(latents_name=name, latents=latents, seed=None)
+noise_output = NoiseOutput(
+    noise=LatentsField(latents_name=latents_name, seed=seed),
+    width=latents.size()[3] * 8,
+    height=latents.size()[2] * 8,
+)
+cond_output = ConditioningOutput(
+    conditioning=ConditioningField(
+        conditioning_name=conditioning_name,
+    ),
+)
+
+# New
+image_output = ImageOutput.build(image_dto)
+latents_output = LatentsOutput.build(latents_name=name, latents=noise, seed=self.seed)
+noise_output = NoiseOutput.build(latents_name=name, latents=noise, seed=self.seed)
+cond_output = ConditioningOutput.build(conditioning_name)
+```
+
+You can still create the objects using constructors if you want, but we suggest
+using the builder methods.

docs/nodes/communityNodes.md

@@ -32,6 +32,7 @@ To use a community workflow, download the the `.json` node graph file and load i
     + [Image to Character Art Image Nodes](#image-to-character-art-image-nodes)
     + [Image Picker](#image-picker)
     + [Image Resize Plus](#image-resize-plus)
+    + [Latent Upscale](#latent-upscale)
     + [Load Video Frame](#load-video-frame)
     + [Make 3D](#make-3d)
     + [Mask Operations](#mask-operations)
@@ -290,6 +291,13 @@ View:
 </br><img src="https://raw.githubusercontent.com/VeyDlin/image-resize-plus-node/master/.readme/node.png" width="500" />
 
 
+--------------------------------
+### Latent Upscale
+
+**Description:** This node uses a small (~2.4mb) model to upscale the latents used in a Stable Diffusion 1.5 or Stable Diffusion XL image generation, rather than the typical interpolation method, avoiding the traditional downsides of the latent upscale technique.
+
+**Node Link:** [https://github.com/gogurtenjoyer/latent-upscale](https://github.com/gogurtenjoyer/latent-upscale)
+
 --------------------------------
 ### Load Video Frame
 
@@ -346,12 +354,21 @@ See full docs here: https://github.com/skunkworxdark/Prompt-tools-nodes/edit/mai
 
 **Description:** A set of nodes for Metadata. Collect Metadata from within an `iterate` node & extract metadata from an image.
 
-- `Metadata Item Linked` - Allows collecting of metadata while within an iterate node with no need for a collect node or conversion to metadata node.
-- `Metadata From Image` - Provides Metadata from an image.
-- `Metadata To String` - Extracts a String value of a label from metadata.
-- `Metadata To Integer` - Extracts an Integer value of a label from metadata.
-- `Metadata To Float` - Extracts a Float value of a label from metadata.
-- `Metadata To Scheduler` - Extracts a Scheduler value of a label from metadata.
+- `Metadata Item Linked` - Allows collecting of metadata while within an iterate node with no need for a collect node or conversion to metadata node
+- `Metadata From Image` - Provides Metadata from an image
+- `Metadata To String` - Extracts a String value of a label from metadata
+- `Metadata To Integer` - Extracts an Integer value of a label from metadata
+- `Metadata To Float` - Extracts a Float value of a label from metadata
+- `Metadata To Scheduler` - Extracts a Scheduler value of a label from metadata
+- `Metadata To Bool` - Extracts Bool types from metadata
+- `Metadata To Model` - Extracts model types from metadata
+- `Metadata To SDXL Model` - Extracts SDXL model types from metadata
+- `Metadata To LoRAs` - Extracts Loras from metadata. 
+- `Metadata To SDXL LoRAs` - Extracts SDXL Loras from metadata
+- `Metadata To ControlNets` - Extracts ControNets from metadata
+- `Metadata To IP-Adapters` - Extracts IP-Adapters from metadata
+- `Metadata To T2I-Adapters` - Extracts T2I-Adapters from metadata
+- `Denoise Latents + Metadata` - This is an inherited version of the existing `Denoise Latents` node but with a metadata input and output. 
 
 **Node Link:** https://github.com/skunkworxdark/metadata-linked-nodes
 

docs/nodes/defaultNodes.md

@@ -19,6 +19,8 @@ their descriptions.
 | Conditioning Primitive                                        | A conditioning tensor primitive value                                                                                                                |
 | Content Shuffle Processor                                     | Applies content shuffle processing to image                                                                                                          |
 | ControlNet                                                    | Collects ControlNet info to pass to other nodes                                                                                                      |
+| Create Denoise Mask                                           | Converts a greyscale or transparency image into a mask for denoising.                                                                                |
+| Create Gradient Mask                                          | Creates a mask for Gradient ("soft", "differential") inpainting that gradually expands during denoising. Improves edge coherence.                    |
 | Denoise Latents                                               | Denoises noisy latents to decodable images                                                                                                           |
 | Divide Integers                                               | Divides two numbers                                                                                                                                  |
 | Dynamic Prompt                                                | Parses a prompt using adieyal/dynamicprompts' random or combinatorial generator                                                                      |

docs/requirements-mkdocs.txt

@@ -1,5 +0,0 @@
-mkdocs
-mkdocs-material>=8, <9
-mkdocs-git-revision-date-localized-plugin
-mkdocs-redirects==1.2.0
-

docs/stylesheets/extra.css

@@ -1,5 +0,0 @@
-:root {
-    --md-primary-fg-color:        #35A4DB;
-    --md-primary-fg-color--light: #35A4DB;
-    --md-primary-fg-color--dark:  #35A4DB;
-  }

installer/create_installer.sh

@@ -2,22 +2,18 @@
 
 set -e
 
-BCYAN="\e[1;36m"
-BYELLOW="\e[1;33m"
-BGREEN="\e[1;32m"
-BRED="\e[1;31m"
-RED="\e[31m"
-RESET="\e[0m"
-
-function is_bin_in_path {
-    builtin type -P "$1" &>/dev/null
-}
+BCYAN="\033[1;36m"
+BYELLOW="\033[1;33m"
+BGREEN="\033[1;32m"
+BRED="\033[1;31m"
+RED="\033[31m"
+RESET="\033[0m"
 
 function git_show {
     git show -s --format=oneline --abbrev-commit "$1" | cat
 }
 
-if [[ -v "VIRTUAL_ENV" ]]; then
+if [[ ! -z "${VIRTUAL_ENV}" ]]; then
     # we can't just call 'deactivate' because this function is not exported
     # to the environment of this script from the bash process that runs the script
     echo -e "${BRED}A virtual environment is activated. Please deactivate it before proceeding.${RESET}"
@@ -26,31 +22,63 @@ fi
 
 cd "$(dirname "$0")"
 
-echo
-echo -e "${BYELLOW}This script must be run from the installer directory!${RESET}"
-echo "The current working directory is $(pwd)"
-read -p "If that looks right, press any key to proceed, or CTRL-C to exit..."
-echo
-
-# Some machines only have `python3` in PATH, others have `python` - make an alias.
-# We can use a function to approximate an alias within a non-interactive shell.
-if ! is_bin_in_path python && is_bin_in_path python3; then
-    function python {
-        python3 "$@"
-    }
-fi
-
 VERSION=$(
     cd ..
-    python -c "from invokeai.version import __version__ as version; print(version)"
+    python3 -c "from invokeai.version import __version__ as version; print(version)"
 )
-PATCH=""
-VERSION="v${VERSION}${PATCH}"
+VERSION="v${VERSION}"
+
+if [[ ! -z ${CI} ]]; then
+    echo
+    echo -e "${BCYAN}CI environment detected${RESET}"
+    echo
+else
+    echo
+    echo -e "${BYELLOW}This script must be run from the installer directory!${RESET}"
+    echo "The current working directory is $(pwd)"
+    read -p "If that looks right, press any key to proceed, or CTRL-C to exit..."
+    echo
+fi
 
 echo -e "${BGREEN}HEAD${RESET}:"
 git_show HEAD
 echo
 
+# ---------------------- FRONTEND ----------------------
+
+pushd ../invokeai/frontend/web >/dev/null
+echo "Installing frontend dependencies..."
+echo
+pnpm i --frozen-lockfile
+echo
+if [[ ! -z ${CI} ]]; then
+    echo "Building frontend without checks..."
+    # In CI, we have already done the frontend checks and can just build
+    pnpm vite build
+else
+    echo "Running checks and building frontend..."
+    # This runs all the frontend checks and builds
+    pnpm build
+fi
+echo
+popd
+
+# ---------------------- BACKEND ----------------------
+
+echo
+echo "Building wheel..."
+echo
+
+# install the 'build' package in the user site packages, if needed
+# could be improved by using a temporary venv, but it's tiny and harmless
+if [[ $(python3 -c 'from importlib.util import find_spec; print(find_spec("build") is None)') == "True" ]]; then
+    pip install --user build
+fi
+
+rm -rf ../build
+
+python3 -m build --outdir dist/ ../.
+
 # ----------------------
 
 echo
@@ -78,10 +106,28 @@ chmod a+x InvokeAI-Installer/install.sh
 cp install.bat.in InvokeAI-Installer/install.bat
 cp WinLongPathsEnabled.reg InvokeAI-Installer/
 
-# Zip everything up
-zip -r InvokeAI-installer-$VERSION.zip InvokeAI-Installer
+FILENAME=InvokeAI-installer-$VERSION.zip
 
-# clean up
-rm -rf InvokeAI-Installer tmp dist ../invokeai/frontend/web/dist/
+# Zip everything up
+zip -r ${FILENAME} InvokeAI-Installer
+
+echo
+echo -e "${BGREEN}Built installer: ./${FILENAME}${RESET}"
+echo -e "${BGREEN}Built PyPi distribution: ./dist${RESET}"
+
+# clean up, but only if we are not in a github action
+if [[ -z ${CI} ]]; then
+    echo
+    echo "Cleaning up intermediate build files..."
+    rm -rf InvokeAI-Installer tmp ../invokeai/frontend/web/dist/
+fi
+
+if [[ ! -z ${CI} ]]; then
+    echo
+    echo "Setting GitHub action outputs..."
+    echo "INSTALLER_FILENAME=${FILENAME}" >>$GITHUB_OUTPUT
+    echo "INSTALLER_PATH=installer/${FILENAME}" >>$GITHUB_OUTPUT
+    echo "DIST_PATH=installer/dist/" >>$GITHUB_OUTPUT
+fi
 
 exit 0

installer/tag_release.sh

@@ -2,12 +2,12 @@
 
 set -e
 
-BCYAN="\e[1;36m"
-BYELLOW="\e[1;33m"
-BGREEN="\e[1;32m"
-BRED="\e[1;31m"
-RED="\e[31m"
-RESET="\e[0m"
+BCYAN="\033[1;36m"
+BYELLOW="\033[1;33m"
+BGREEN="\033[1;32m"
+BRED="\033[1;31m"
+RED="\033[31m"
+RESET="\033[0m"
 
 function does_tag_exist {
     git rev-parse --quiet --verify "refs/tags/$1" >/dev/null
@@ -23,49 +23,40 @@ function git_show {
 
 VERSION=$(
     cd ..
-    python -c "from invokeai.version import __version__ as version; print(version)"
+    python3 -c "from invokeai.version import __version__ as version; print(version)"
 )
 PATCH=""
-MAJOR_VERSION=$(echo $VERSION | sed 's/\..*$//')
 VERSION="v${VERSION}${PATCH}"
-LATEST_TAG="v${MAJOR_VERSION}-latest"
 
 if does_tag_exist $VERSION; then
     echo -e "${BCYAN}${VERSION}${RESET} already exists:"
     git_show_ref tags/$VERSION
     echo
 fi
-if does_tag_exist $LATEST_TAG; then
-    echo -e "${BCYAN}${LATEST_TAG}${RESET} already exists:"
-    git_show_ref tags/$LATEST_TAG
-    echo
-fi
 
 echo -e "${BGREEN}HEAD${RESET}:"
 git_show
 echo
 
-echo -e -n "Create tags ${BCYAN}${VERSION}${RESET} and ${BCYAN}${LATEST_TAG}${RESET} @ ${BGREEN}HEAD${RESET}, ${RED}deleting existing tags on remote${RESET}? "
+echo -e "${BGREEN}git remote -v${RESET}:"
+git remote -v
+echo
+
+echo -e -n "Create tags ${BCYAN}${VERSION}${RESET} @ ${BGREEN}HEAD${RESET}, ${RED}deleting existing tags on origin remote${RESET}? "
 read -e -p 'y/n [n]: ' input
 RESPONSE=${input:='n'}
 if [ "$RESPONSE" == 'y' ]; then
     echo
-    echo -e "Deleting ${BCYAN}${VERSION}${RESET} tag on remote..."
-    git push --delete origin $VERSION
+    echo -e "Deleting ${BCYAN}${VERSION}${RESET} tag on origin remote..."
+    git push origin :refs/tags/$VERSION
 
-    echo -e "Tagging ${BGREEN}HEAD${RESET} with ${BCYAN}${VERSION}${RESET} locally..."
+    echo -e "Tagging ${BGREEN}HEAD${RESET} with ${BCYAN}${VERSION}${RESET} on locally..."
     if ! git tag -fa $VERSION; then
         echo "Existing/invalid tag"
         exit -1
     fi
 
-    echo -e "Deleting ${BCYAN}${LATEST_TAG}${RESET} tag on remote..."
-    git push --delete origin $LATEST_TAG
-
-    echo -e "Tagging ${BGREEN}HEAD${RESET} with ${BCYAN}${LATEST_TAG}${RESET} locally..."
-    git tag -fa $LATEST_TAG
-
-    echo -e "Pushing updated tags to remote..."
+    echo -e "Pushing updated tags to origin remote..."
     git push origin --tags
 fi
 exit 0

invokeai/app/api/dependencies.py

@@ -25,8 +25,8 @@ from ..services.invocation_cache.invocation_cache_memory import MemoryInvocation
 from ..services.invocation_services import InvocationServices
 from ..services.invocation_stats.invocation_stats_default import InvocationStatsService
 from ..services.invoker import Invoker
+from ..services.model_images.model_images_default import ModelImageFileStorageDisk
 from ..services.model_manager.model_manager_default import ModelManagerService
-from ..services.model_metadata import ModelMetadataStoreSQL
 from ..services.model_records import ModelRecordServiceSQL
 from ..services.names.names_default import SimpleNameService
 from ..services.session_processor.session_processor_default import DefaultSessionProcessor
@@ -72,6 +72,8 @@ class ApiDependencies:
 
         image_files = DiskImageFileStorage(f"{output_folder}/images")
 
+        model_images_folder = config.models_path
+
         db = init_db(config=config, logger=logger, image_files=image_files)
 
         configuration = config
@@ -93,10 +95,10 @@ class ApiDependencies:
             ObjectSerializerDisk[ConditioningFieldData](output_folder / "conditioning", ephemeral=True)
         )
         download_queue_service = DownloadQueueService(event_bus=events)
-        model_metadata_service = ModelMetadataStoreSQL(db=db)
+        model_images_service = ModelImageFileStorageDisk(model_images_folder / "model_images")
         model_manager = ModelManagerService.build_model_manager(
             app_config=configuration,
-            model_record_service=ModelRecordServiceSQL(db=db, metadata_store=model_metadata_service),
+            model_record_service=ModelRecordServiceSQL(db=db),
             download_queue=download_queue_service,
             events=events,
         )
@@ -120,6 +122,7 @@ class ApiDependencies:
             images=images,
             invocation_cache=invocation_cache,
             logger=logger,
+            model_images=model_images_service,
             model_manager=model_manager,
             download_queue=download_queue_service,
             names=names,

invokeai/app/api/routers/model_manager.py

@@ -1,27 +1,26 @@
 # Copyright (c) 2023 Lincoln D. Stein
 """FastAPI route for model configuration records."""
 
+import io
 import pathlib
 import shutil
-from hashlib import sha1
-from random import randbytes
-from typing import Any, Dict, List, Optional, Set
+import traceback
+from typing import Any, Dict, List, Optional
 
-from fastapi import Body, Path, Query, Response
+from fastapi import Body, Path, Query, Response, UploadFile
+from fastapi.responses import FileResponse
 from fastapi.routing import APIRouter
-from pydantic import BaseModel, ConfigDict, Field
+from PIL import Image
+from pydantic import AnyHttpUrl, BaseModel, ConfigDict, Field
 from starlette.exceptions import HTTPException
 from typing_extensions import Annotated
 
 from invokeai.app.services.model_install import ModelInstallJob
 from invokeai.app.services.model_records import (
-    DuplicateModelException,
     InvalidModelException,
-    ModelRecordOrderBy,
-    ModelSummary,
     UnknownModelException,
 )
-from invokeai.app.services.shared.pagination import PaginatedResults
+from invokeai.app.services.model_records.model_records_base import DuplicateModelException, ModelRecordChanges
 from invokeai.backend.model_manager.config import (
     AnyModelConfig,
     BaseModelType,
@@ -30,14 +29,17 @@ from invokeai.backend.model_manager.config import (
     ModelType,
     SubModelType,
 )
-from invokeai.backend.model_manager.merge import MergeInterpolationMethod, ModelMerger
-from invokeai.backend.model_manager.metadata import AnyModelRepoMetadata
+from invokeai.backend.model_manager.metadata.fetch.huggingface import HuggingFaceMetadataFetch
+from invokeai.backend.model_manager.metadata.metadata_base import ModelMetadataWithFiles, UnknownMetadataException
 from invokeai.backend.model_manager.search import ModelSearch
 
 from ..dependencies import ApiDependencies
 
 model_manager_router = APIRouter(prefix="/v2/models", tags=["model_manager"])
 
+# images are immutable; set a high max-age
+IMAGE_MAX_AGE = 31536000
+
 
 class ModelsList(BaseModel):
     """Return list of configs."""
@@ -47,15 +49,6 @@ class ModelsList(BaseModel):
     model_config = ConfigDict(use_enum_values=True)
 
 
-class ModelTagSet(BaseModel):
-    """Return tags for a set of models."""
-
-    key: str
-    name: str
-    author: str
-    tags: Set[str]
-
-
 ##############################################################################
 # These are example inputs and outputs that are used in places where Swagger
 # is unable to generate a correct example.
@@ -66,19 +59,16 @@ example_model_config = {
     "base": "sd-1",
     "type": "main",
     "format": "checkpoint",
-    "config": "string",
+    "config_path": "string",
     "key": "string",
-    "original_hash": "string",
-    "current_hash": "string",
+    "hash": "string",
     "description": "string",
     "source": "string",
-    "last_modified": 0,
-    "vae": "string",
+    "converted_at": 0,
     "variant": "normal",
     "prediction_type": "epsilon",
     "repo_variant": "fp16",
     "upcast_attention": False,
-    "ztsnr_training": False,
 }
 
 example_model_input = {
@@ -87,50 +77,12 @@ example_model_input = {
     "base": "sd-1",
     "type": "main",
     "format": "checkpoint",
-    "config": "configs/stable-diffusion/v1-inference.yaml",
+    "config_path": "configs/stable-diffusion/v1-inference.yaml",
     "description": "Model description",
     "vae": None,
     "variant": "normal",
 }
 
-example_model_metadata = {
-    "name": "ip_adapter_sd_image_encoder",
-    "author": "InvokeAI",
-    "tags": [
-        "transformers",
-        "safetensors",
-        "clip_vision_model",
-        "endpoints_compatible",
-        "region:us",
-        "has_space",
-        "license:apache-2.0",
-    ],
-    "files": [
-        {
-            "url": "https://huggingface.co/InvokeAI/ip_adapter_sd_image_encoder/resolve/main/README.md",
-            "path": "ip_adapter_sd_image_encoder/README.md",
-            "size": 628,
-            "sha256": None,
-        },
-        {
-            "url": "https://huggingface.co/InvokeAI/ip_adapter_sd_image_encoder/resolve/main/config.json",
-            "path": "ip_adapter_sd_image_encoder/config.json",
-            "size": 560,
-            "sha256": None,
-        },
-        {
-            "url": "https://huggingface.co/InvokeAI/ip_adapter_sd_image_encoder/resolve/main/model.safetensors",
-            "path": "ip_adapter_sd_image_encoder/model.safetensors",
-            "size": 2528373448,
-            "sha256": "6ca9667da1ca9e0b0f75e46bb030f7e011f44f86cbfb8d5a36590fcd7507b030",
-        },
-    ],
-    "type": "huggingface",
-    "id": "InvokeAI/ip_adapter_sd_image_encoder",
-    "tag_dict": {"license": "apache-2.0"},
-    "last_modified": "2023-09-23T17:33:25Z",
-}
-
 ##############################################################################
 # ROUTES
 ##############################################################################
@@ -162,6 +114,9 @@ async def list_model_records(
         found_models.extend(
             record_store.search_by_attr(model_type=model_type, model_name=model_name, model_format=model_format)
         )
+    for model in found_models:
+        cover_image = ApiDependencies.invoker.services.model_images.get_url(model.key)
+        model.cover_image = cover_image
     return ModelsList(models=found_models)
 
 
@@ -205,53 +160,23 @@ async def get_model_record(
     record_store = ApiDependencies.invoker.services.model_manager.store
     try:
         config: AnyModelConfig = record_store.get_model(key)
+        cover_image = ApiDependencies.invoker.services.model_images.get_url(key)
+        config.cover_image = cover_image
         return config
     except UnknownModelException as e:
         raise HTTPException(status_code=404, detail=str(e))
 
 
-@model_manager_router.get("/summary", operation_id="list_model_summary")
-async def list_model_summary(
-    page: int = Query(default=0, description="The page to get"),
-    per_page: int = Query(default=10, description="The number of models per page"),
-    order_by: ModelRecordOrderBy = Query(default=ModelRecordOrderBy.Default, description="The attribute to order by"),
-) -> PaginatedResults[ModelSummary]:
-    """Gets a page of model summary data."""
-    record_store = ApiDependencies.invoker.services.model_manager.store
-    results: PaginatedResults[ModelSummary] = record_store.list_models(page=page, per_page=per_page, order_by=order_by)
-    return results
-
-
-@model_manager_router.get(
-    "/i/{key}/metadata",
-    operation_id="get_model_metadata",
-    responses={
-        200: {
-            "description": "The model metadata was retrieved successfully",
-            "content": {"application/json": {"example": example_model_metadata}},
-        },
-        400: {"description": "Bad request"},
-    },
-)
-async def get_model_metadata(
-    key: str = Path(description="Key of the model repo metadata to fetch."),
-) -> Optional[AnyModelRepoMetadata]:
-    """Get a model metadata object."""
-    record_store = ApiDependencies.invoker.services.model_manager.store
-    result: Optional[AnyModelRepoMetadata] = record_store.get_metadata(key)
-
-    return result
-
-
-@model_manager_router.get(
-    "/tags",
-    operation_id="list_tags",
-)
-async def list_tags() -> Set[str]:
-    """Get a unique set of all the model tags."""
-    record_store = ApiDependencies.invoker.services.model_manager.store
-    result: Set[str] = record_store.list_tags()
-    return result
+# @model_manager_router.get("/summary", operation_id="list_model_summary")
+# async def list_model_summary(
+#     page: int = Query(default=0, description="The page to get"),
+#     per_page: int = Query(default=10, description="The number of models per page"),
+#     order_by: ModelRecordOrderBy = Query(default=ModelRecordOrderBy.Default, description="The attribute to order by"),
+# ) -> PaginatedResults[ModelSummary]:
+#     """Gets a page of model summary data."""
+#     record_store = ApiDependencies.invoker.services.model_manager.store
+#     results: PaginatedResults[ModelSummary] = record_store.list_models(page=page, per_page=per_page, order_by=order_by)
+#     return results
 
 
 class FoundModel(BaseModel):
@@ -323,17 +248,38 @@ async def scan_for_models(
     return scan_results
 
 
+class HuggingFaceModels(BaseModel):
+    urls: List[AnyHttpUrl] | None = Field(description="URLs for all checkpoint format models in the metadata")
+    is_diffusers: bool = Field(description="Whether the metadata is for a Diffusers format model")
+
+
 @model_manager_router.get(
-    "/tags/search",
-    operation_id="search_by_metadata_tags",
+    "/hugging_face",
+    operation_id="get_hugging_face_models",
+    responses={
+        200: {"description": "Hugging Face repo scanned successfully"},
+        400: {"description": "Invalid hugging face repo"},
+    },
+    status_code=200,
+    response_model=HuggingFaceModels,
 )
-async def search_by_metadata_tags(
-    tags: Set[str] = Query(default=None, description="Tags to search for"),
-) -> ModelsList:
-    """Get a list of models."""
-    record_store = ApiDependencies.invoker.services.model_manager.store
-    results = record_store.search_by_metadata_tag(tags)
-    return ModelsList(models=results)
+async def get_hugging_face_models(
+    hugging_face_repo: str = Query(description="Hugging face repo to search for models", default=None),
+) -> HuggingFaceModels:
+    try:
+        metadata = HuggingFaceMetadataFetch().from_id(hugging_face_repo)
+    except UnknownMetadataException:
+        raise HTTPException(
+            status_code=400,
+            detail="No HuggingFace repository found",
+        )
+
+    assert isinstance(metadata, ModelMetadataWithFiles)
+
+    return HuggingFaceModels(
+        urls=metadata.ckpt_urls,
+        is_diffusers=metadata.is_diffusers,
+    )
 
 
 @model_manager_router.patch(
@@ -352,15 +298,13 @@ async def search_by_metadata_tags(
 )
 async def update_model_record(
     key: Annotated[str, Path(description="Unique key of model")],
-    info: Annotated[
-        AnyModelConfig, Body(description="Model config", discriminator="type", example=example_model_input)
-    ],
+    changes: Annotated[ModelRecordChanges, Body(description="Model config", example=example_model_input)],
 ) -> AnyModelConfig:
-    """Update model contents with a new config. If the model name or base fields are changed, then the model is renamed."""
+    """Update a model's config."""
     logger = ApiDependencies.invoker.services.logger
     record_store = ApiDependencies.invoker.services.model_manager.store
     try:
-        model_response: AnyModelConfig = record_store.update_model(key, config=info)
+        model_response: AnyModelConfig = record_store.update_model(key, changes=changes)
         logger.info(f"Updated model: {key}")
     except UnknownModelException as e:
         raise HTTPException(status_code=404, detail=str(e))
@@ -370,16 +314,85 @@ async def update_model_record(
     return model_response
 
 
+@model_manager_router.get(
+    "/i/{key}/image",
+    operation_id="get_model_image",
+    responses={
+        200: {
+            "description": "The model image was fetched successfully",
+        },
+        400: {"description": "Bad request"},
+        404: {"description": "The model image could not be found"},
+    },
+    status_code=200,
+)
+async def get_model_image(
+    key: str = Path(description="The name of model image file to get"),
+) -> FileResponse:
+    """Gets an image file that previews the model"""
+
+    try:
+        path = ApiDependencies.invoker.services.model_images.get_path(key)
+
+        response = FileResponse(
+            path,
+            media_type="image/png",
+            filename=key + ".png",
+            content_disposition_type="inline",
+        )
+        response.headers["Cache-Control"] = f"max-age={IMAGE_MAX_AGE}"
+        return response
+    except Exception:
+        raise HTTPException(status_code=404)
+
+
+@model_manager_router.patch(
+    "/i/{key}/image",
+    operation_id="update_model_image",
+    responses={
+        200: {
+            "description": "The model image was updated successfully",
+        },
+        400: {"description": "Bad request"},
+    },
+    status_code=200,
+)
+async def update_model_image(
+    key: Annotated[str, Path(description="Unique key of model")],
+    image: UploadFile,
+) -> None:
+    if not image.content_type or not image.content_type.startswith("image"):
+        raise HTTPException(status_code=415, detail="Not an image")
+
+    contents = await image.read()
+    try:
+        pil_image = Image.open(io.BytesIO(contents))
+
+    except Exception:
+        ApiDependencies.invoker.services.logger.error(traceback.format_exc())
+        raise HTTPException(status_code=415, detail="Failed to read image")
+
+    logger = ApiDependencies.invoker.services.logger
+    model_images = ApiDependencies.invoker.services.model_images
+    try:
+        model_images.save(pil_image, key)
+        logger.info(f"Updated image for model: {key}")
+    except ValueError as e:
+        logger.error(str(e))
+        raise HTTPException(status_code=409, detail=str(e))
+    return
+
+
 @model_manager_router.delete(
     "/i/{key}",
-    operation_id="del_model_record",
+    operation_id="delete_model",
     responses={
         204: {"description": "Model deleted successfully"},
         404: {"description": "Model not found"},
     },
     status_code=204,
 )
-async def del_model_record(
+async def delete_model(
     key: str = Path(description="Unique key of model to remove from model registry."),
 ) -> Response:
     """
@@ -400,42 +413,62 @@ async def del_model_record(
         raise HTTPException(status_code=404, detail=str(e))
 
 
-@model_manager_router.post(
-    "/i/",
-    operation_id="add_model_record",
+@model_manager_router.delete(
+    "/i/{key}/image",
+    operation_id="delete_model_image",
     responses={
-        201: {
-            "description": "The model added successfully",
-            "content": {"application/json": {"example": example_model_config}},
+        204: {"description": "Model image deleted successfully"},
+        404: {"description": "Model image not found"},
     },
-        409: {"description": "There is already a model corresponding to this path or repo_id"},
-        415: {"description": "Unrecognized file/folder format"},
-    },
-    status_code=201,
+    status_code=204,
 )
-async def add_model_record(
-    config: Annotated[
-        AnyModelConfig, Body(description="Model config", discriminator="type", example=example_model_input)
-    ],
-) -> AnyModelConfig:
-    """Add a model using the configuration information appropriate for its type."""
+async def delete_model_image(
+    key: str = Path(description="Unique key of model image to remove from model_images directory."),
+) -> None:
     logger = ApiDependencies.invoker.services.logger
-    record_store = ApiDependencies.invoker.services.model_manager.store
-    if config.key == "<NOKEY>":
-        config.key = sha1(randbytes(100)).hexdigest()
-        logger.info(f"Created model {config.key} for {config.name}")
+    model_images = ApiDependencies.invoker.services.model_images
     try:
-        record_store.add_model(config.key, config)
-    except DuplicateModelException as e:
+        model_images.delete(key)
+        logger.info(f"Deleted model image: {key}")
+        return
+    except UnknownModelException as e:
         logger.error(str(e))
-        raise HTTPException(status_code=409, detail=str(e))
-    except InvalidModelException as e:
-        logger.error(str(e))
-        raise HTTPException(status_code=415)
+        raise HTTPException(status_code=404, detail=str(e))
 
-    # now fetch it out
-    result: AnyModelConfig = record_store.get_model(config.key)
-    return result
+
+# @model_manager_router.post(
+#     "/i/",
+#     operation_id="add_model_record",
+#     responses={
+#         201: {
+#             "description": "The model added successfully",
+#             "content": {"application/json": {"example": example_model_config}},
+#         },
+#         409: {"description": "There is already a model corresponding to this path or repo_id"},
+#         415: {"description": "Unrecognized file/folder format"},
+#     },
+#     status_code=201,
+# )
+# async def add_model_record(
+#     config: Annotated[
+#         AnyModelConfig, Body(description="Model config", discriminator="type", example=example_model_input)
+#     ],
+# ) -> AnyModelConfig:
+#     """Add a model using the configuration information appropriate for its type."""
+#     logger = ApiDependencies.invoker.services.logger
+#     record_store = ApiDependencies.invoker.services.model_manager.store
+#     try:
+#         record_store.add_model(config)
+#     except DuplicateModelException as e:
+#         logger.error(str(e))
+#         raise HTTPException(status_code=409, detail=str(e))
+#     except InvalidModelException as e:
+#         logger.error(str(e))
+#         raise HTTPException(status_code=415)
+
+#     # now fetch it out
+#     result: AnyModelConfig = record_store.get_model(config.key)
+#     return result
 
 
 @model_manager_router.post(
@@ -451,6 +484,7 @@ async def add_model_record(
 )
 async def install_model(
     source: str = Query(description="Model source to install, can be a local path, repo_id, or remote URL"),
+    inplace: Optional[bool] = Query(description="Whether or not to install a local model in place", default=False),
     # TODO(MM2): Can we type this?
     config: Optional[Dict[str, Any]] = Body(
         description="Dict of fields that override auto-probed values in the model config record, such as name, description and prediction_type ",
@@ -493,6 +527,7 @@ async def install_model(
             source=source,
             config=config,
             access_token=access_token,
+            inplace=bool(inplace),
         )
         logger.info(f"Started installation of {source}")
     except UnknownModelException as e:
@@ -508,10 +543,10 @@ async def install_model(
 
 
 @model_manager_router.get(
-    "/import",
-    operation_id="list_model_install_jobs",
+    "/install",
+    operation_id="list_model_installs",
 )
-async def list_model_install_jobs() -> List[ModelInstallJob]:
+async def list_model_installs() -> List[ModelInstallJob]:
     """Return the list of model install jobs.
 
     Install jobs have a numeric `id`, a `status`, and other fields that provide information on
@@ -525,9 +560,8 @@ async def list_model_install_jobs() -> List[ModelInstallJob]:
     * "cancelled" -- Job was cancelled before completion.
 
     Once completed, information about the model such as its size, base
-    model, type, and metadata can be retrieved from the `config_out`
-    field. For multi-file models such as diffusers, information on individual files
-    can be retrieved from `download_parts`.
+    model and type can be retrieved from the `config_out` field. For multi-file models such as diffusers,
+    information on individual files can be retrieved from `download_parts`.
 
     See the example and schema below for more information.
     """
@@ -536,7 +570,7 @@ async def list_model_install_jobs() -> List[ModelInstallJob]:
 
 
 @model_manager_router.get(
-    "/import/{id}",
+    "/install/{id}",
     operation_id="get_model_install_job",
     responses={
         200: {"description": "Success"},
@@ -556,7 +590,7 @@ async def get_model_install_job(id: int = Path(description="Model install id"))
 
 
 @model_manager_router.delete(
-    "/import/{id}",
+    "/install/{id}",
     operation_id="cancel_model_install_job",
     responses={
         201: {"description": "The job was cancelled successfully"},
@@ -574,8 +608,8 @@ async def cancel_model_install_job(id: int = Path(description="Model install job
     installer.cancel_job(job)
 
 
-@model_manager_router.patch(
-    "/import",
+@model_manager_router.delete(
+    "/install",
     operation_id="prune_model_install_jobs",
     responses={
         204: {"description": "All completed and errored jobs have been pruned"},
@@ -645,7 +679,7 @@ async def convert_model(
         raise HTTPException(400, f"The model with key {key} is not a main checkpoint model.")
 
     # loading the model will convert it into a cached diffusers file
-    model_manager.load_model_by_config(model_config, submodel_type=SubModelType.Scheduler)
+    model_manager.load.load_model(model_config, submodel_type=SubModelType.Scheduler)
 
     # Get the path of the converted model from the loader
     cache_path = loader.convert_cache.cache_path(key)
@@ -654,7 +688,8 @@ async def convert_model(
     # temporarily rename the original safetensors file so that there is no naming conflict
     original_name = model_config.name
     model_config.name = f"{original_name}.DELETE"
-    store.update_model(key, config=model_config)
+    changes = ModelRecordChanges(name=model_config.name)
+    store.update_model(key, changes=changes)
 
     # install the diffusers
     try:
@@ -663,7 +698,7 @@ async def convert_model(
             config={
                 "name": original_name,
                 "description": model_config.description,
-                "original_hash": model_config.original_hash,
+                "hash": model_config.hash,
                 "source": model_config.source,
             },
         )
@@ -671,10 +706,6 @@ async def convert_model(
         logger.error(str(e))
         raise HTTPException(status_code=409, detail=str(e))
 
-    # get the original metadata
-    if orig_metadata := store.get_metadata(key):
-        store.metadata_store.add_metadata(new_key, orig_metadata)
-
     # delete the original safetensors file
     installer.delete(key)
 
@@ -686,66 +717,66 @@ async def convert_model(
     return new_config
 
 
-@model_manager_router.put(
-    "/merge",
-    operation_id="merge",
-    responses={
-        200: {
-            "description": "Model converted successfully",
-            "content": {"application/json": {"example": example_model_config}},
-        },
-        400: {"description": "Bad request"},
-        404: {"description": "Model not found"},
-        409: {"description": "There is already a model registered at this location"},
-    },
-)
-async def merge(
-    keys: List[str] = Body(description="Keys for two to three models to merge", min_length=2, max_length=3),
-    merged_model_name: Optional[str] = Body(description="Name of destination model", default=None),
-    alpha: float = Body(description="Alpha weighting strength to apply to 2d and 3d models", default=0.5),
-    force: bool = Body(
-        description="Force merging of models created with different versions of diffusers",
-        default=False,
-    ),
-    interp: Optional[MergeInterpolationMethod] = Body(description="Interpolation method", default=None),
-    merge_dest_directory: Optional[str] = Body(
-        description="Save the merged model to the designated directory (with 'merged_model_name' appended)",
-        default=None,
-    ),
-) -> AnyModelConfig:
-    """
-    Merge diffusers models. The process is controlled by a set parameters provided in the body of the request.
-    ```
-    Argument                Description [default]
-    --------               ----------------------
-    keys                   List of 2-3 model keys to merge together. All models must use the same base type.
-    merged_model_name      Name for the merged model [Concat model names]
-    alpha                  Alpha value (0.0-1.0). Higher values give more weight to the second model [0.5]
-    force                  If true, force the merge even if the models were generated by different versions of the diffusers library [False]
-    interp                 Interpolation method. One of "weighted_sum", "sigmoid", "inv_sigmoid" or "add_difference" [weighted_sum]
-    merge_dest_directory   Specify a directory to store the merged model in [models directory]
-    ```
-    """
-    logger = ApiDependencies.invoker.services.logger
-    try:
-        logger.info(f"Merging models: {keys} into {merge_dest_directory or '<MODELS>'}/{merged_model_name}")
-        dest = pathlib.Path(merge_dest_directory) if merge_dest_directory else None
-        installer = ApiDependencies.invoker.services.model_manager.install
-        merger = ModelMerger(installer)
-        model_names = [installer.record_store.get_model(x).name for x in keys]
-        response = merger.merge_diffusion_models_and_save(
-            model_keys=keys,
-            merged_model_name=merged_model_name or "+".join(model_names),
-            alpha=alpha,
-            interp=interp,
-            force=force,
-            merge_dest_directory=dest,
-        )
-    except UnknownModelException:
-        raise HTTPException(
-            status_code=404,
-            detail=f"One or more of the models '{keys}' not found",
-        )
-    except ValueError as e:
-        raise HTTPException(status_code=400, detail=str(e))
-    return response
+# @model_manager_router.put(
+#     "/merge",
+#     operation_id="merge",
+#     responses={
+#         200: {
+#             "description": "Model converted successfully",
+#             "content": {"application/json": {"example": example_model_config}},
+#         },
+#         400: {"description": "Bad request"},
+#         404: {"description": "Model not found"},
+#         409: {"description": "There is already a model registered at this location"},
+#     },
+# )
+# async def merge(
+#     keys: List[str] = Body(description="Keys for two to three models to merge", min_length=2, max_length=3),
+#     merged_model_name: Optional[str] = Body(description="Name of destination model", default=None),
+#     alpha: float = Body(description="Alpha weighting strength to apply to 2d and 3d models", default=0.5),
+#     force: bool = Body(
+#         description="Force merging of models created with different versions of diffusers",
+#         default=False,
+#     ),
+#     interp: Optional[MergeInterpolationMethod] = Body(description="Interpolation method", default=None),
+#     merge_dest_directory: Optional[str] = Body(
+#         description="Save the merged model to the designated directory (with 'merged_model_name' appended)",
+#         default=None,
+#     ),
+# ) -> AnyModelConfig:
+#     """
+#     Merge diffusers models. The process is controlled by a set parameters provided in the body of the request.
+#     ```
+#     Argument                Description [default]
+#     --------               ----------------------
+#     keys                   List of 2-3 model keys to merge together. All models must use the same base type.
+#     merged_model_name      Name for the merged model [Concat model names]
+#     alpha                  Alpha value (0.0-1.0). Higher values give more weight to the second model [0.5]
+#     force                  If true, force the merge even if the models were generated by different versions of the diffusers library [False]
+#     interp                 Interpolation method. One of "weighted_sum", "sigmoid", "inv_sigmoid" or "add_difference" [weighted_sum]
+#     merge_dest_directory   Specify a directory to store the merged model in [models directory]
+#     ```
+#     """
+#     logger = ApiDependencies.invoker.services.logger
+#     try:
+#         logger.info(f"Merging models: {keys} into {merge_dest_directory or '<MODELS>'}/{merged_model_name}")
+#         dest = pathlib.Path(merge_dest_directory) if merge_dest_directory else None
+#         installer = ApiDependencies.invoker.services.model_manager.install
+#         merger = ModelMerger(installer)
+#         model_names = [installer.record_store.get_model(x).name for x in keys]
+#         response = merger.merge_diffusion_models_and_save(
+#             model_keys=keys,
+#             merged_model_name=merged_model_name or "+".join(model_names),
+#             alpha=alpha,
+#             interp=interp,
+#             force=force,
+#             merge_dest_directory=dest,
+#         )
+#     except UnknownModelException:
+#         raise HTTPException(
+#             status_code=404,
+#             detail=f"One or more of the models '{keys}' not found",
+#         )
+#     except ValueError as e:
+#         raise HTTPException(status_code=400, detail=str(e))
+#     return response

invokeai/app/api_app.py

@@ -3,10 +3,10 @@
 # values from the command line or config file.
 import sys
 
-from invokeai.app.api.no_cache_staticfiles import NoCacheStaticFiles
+from invokeai.app.invocations.model import ModelIdentifierField
+from invokeai.app.services.session_processor.session_processor_common import ProgressImage
 from invokeai.version.invokeai_version import __version__
 
-from .invocations.fields import InputFieldJSONSchemaExtra, OutputFieldJSONSchemaExtra
 from .services.config import InvokeAIAppConfig
 
 app_config = InvokeAIAppConfig.get_config()
@@ -19,6 +19,7 @@ if True:  # hack to make flake8 happy with imports coming after setting up the c
     import asyncio
     import mimetypes
     import socket
+    from contextlib import asynccontextmanager
     from inspect import signature
     from pathlib import Path
     from typing import Any
@@ -39,6 +40,7 @@ if True:  # hack to make flake8 happy with imports coming after setting up the c
     # noinspection PyUnresolvedReferences
     import invokeai.backend.util.hotfixes  # noqa: F401 (monkeypatching on import)
     import invokeai.frontend.web as web_dir
+    from invokeai.app.api.no_cache_staticfiles import NoCacheStaticFiles
 
     from ..backend.util.logging import InvokeAILogger
     from .api.dependencies import ApiDependencies
@@ -58,6 +60,7 @@ if True:  # hack to make flake8 happy with imports coming after setting up the c
         BaseInvocation,
         UIConfigBase,
     )
+    from .invocations.fields import InputFieldJSONSchemaExtra, OutputFieldJSONSchemaExtra
 
     if is_mps_available():
         import invokeai.backend.util.mps_fixes  # noqa: F401 (monkeypatching on import)
@@ -71,9 +74,25 @@ logger = InvokeAILogger.get_logger(config=app_config)
 mimetypes.add_type("application/javascript", ".js")
 mimetypes.add_type("text/css", ".css")
 
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    # Add startup event to load dependencies
+    ApiDependencies.initialize(config=app_config, event_handler_id=event_handler_id, logger=logger)
+    yield
+    # Shut down threads
+    ApiDependencies.shutdown()
+
+
 # Create the app
 # TODO: create this all in a method so configuration/etc. can be passed in?
-app = FastAPI(title="Invoke - Community Edition", docs_url=None, redoc_url=None, separate_input_output_schemas=False)
+app = FastAPI(
+    title="Invoke - Community Edition",
+    docs_url=None,
+    redoc_url=None,
+    separate_input_output_schemas=False,
+    lifespan=lifespan,
+)
 
 # Add event handler
 event_handler_id: int = id(app)
@@ -96,18 +115,6 @@ app.add_middleware(
 app.add_middleware(GZipMiddleware, minimum_size=1000)
 
 
-# Add startup event to load dependencies
-@app.on_event("startup")
-async def startup_event() -> None:
-    ApiDependencies.initialize(config=app_config, event_handler_id=event_handler_id, logger=logger)
-
-
-# Shut down threads
-@app.on_event("shutdown")
-async def shutdown_event() -> None:
-    ApiDependencies.shutdown()
-
-
 # Include all routers
 app.include_router(utilities.utilities_router, prefix="/api")
 app.include_router(model_manager.model_manager_router, prefix="/api")
@@ -151,17 +158,19 @@ def custom_openapi() -> dict[str, Any]:
         openapi_schema["components"]["schemas"][schema_key] = output_schema
         openapi_schema["components"]["schemas"][schema_key]["class"] = "output"
 
-    # Add Node Editor UI helper schemas
-    ui_config_schemas = models_json_schema(
+    # Some models don't end up in the schemas as standalone definitions
+    additional_schemas = models_json_schema(
         [
             (UIConfigBase, "serialization"),
             (InputFieldJSONSchemaExtra, "serialization"),
             (OutputFieldJSONSchemaExtra, "serialization"),
+            (ModelIdentifierField, "serialization"),
+            (ProgressImage, "serialization"),
         ],
         ref_template="#/components/schemas/{model}",
     )
-    for schema_key, ui_config_schema in ui_config_schemas[1]["$defs"].items():
-        openapi_schema["components"]["schemas"][schema_key] = ui_config_schema
+    for schema_key, schema_json in additional_schemas[1]["$defs"].items():
+        openapi_schema["components"]["schemas"][schema_key] = schema_json
 
     # Add a reference to the output type to additionalProperties of the invoker schema
     for invoker in all_invocations:

invokeai/app/invocations/compel.py

@@ -1,17 +1,11 @@
-from typing import Iterator, List, Optional, Tuple, Union
+from typing import Iterator, List, Optional, Tuple, Union, cast
 
 import torch
 from compel import Compel, ReturnedEmbeddingsType
 from compel.prompt_parser import Blend, Conjunction, CrossAttentionControlSubstitute, FlattenedPrompt, Fragment
-from transformers import CLIPTextModel, CLIPTokenizer
+from transformers import CLIPTextModel, CLIPTextModelWithProjection, CLIPTokenizer
 
-from invokeai.app.invocations.fields import (
-    FieldDescriptions,
-    Input,
-    InputField,
-    OutputField,
-    UIComponent,
-)
+from invokeai.app.invocations.fields import FieldDescriptions, Input, InputField, OutputField, UIComponent
 from invokeai.app.invocations.primitives import ConditioningOutput
 from invokeai.app.services.shared.invocation_context import InvocationContext
 from invokeai.app.util.ti_utils import generate_ti_list
@@ -25,13 +19,8 @@ from invokeai.backend.stable_diffusion.diffusion.conditioning_data import (
 )
 from invokeai.backend.util.devices import torch_dtype
 
-from .baseinvocation import (
-    BaseInvocation,
-    BaseInvocationOutput,
-    invocation,
-    invocation_output,
-)
-from .model import ClipField
+from .baseinvocation import BaseInvocation, BaseInvocationOutput, invocation, invocation_output
+from .model import CLIPField
 
 # unconditioned: Optional[torch.Tensor]
 
@@ -57,7 +46,7 @@ class CompelInvocation(BaseInvocation):
         description=FieldDescriptions.compel_prompt,
         ui_component=UIComponent.Textarea,
     )
-    clip: ClipField = InputField(
+    clip: CLIPField = InputField(
         title="CLIP",
         description=FieldDescriptions.clip,
         input=Input.Connection,
@@ -65,16 +54,16 @@ class CompelInvocation(BaseInvocation):
 
     @torch.no_grad()
     def invoke(self, context: InvocationContext) -> ConditioningOutput:
-        tokenizer_info = context.models.load(**self.clip.tokenizer.model_dump())
+        tokenizer_info = context.models.load(self.clip.tokenizer)
         tokenizer_model = tokenizer_info.model
         assert isinstance(tokenizer_model, CLIPTokenizer)
-        text_encoder_info = context.models.load(**self.clip.text_encoder.model_dump())
+        text_encoder_info = context.models.load(self.clip.text_encoder)
         text_encoder_model = text_encoder_info.model
         assert isinstance(text_encoder_model, CLIPTextModel)
 
         def _lora_loader() -> Iterator[Tuple[LoRAModelRaw, float]]:
             for lora in self.clip.loras:
-                lora_info = context.models.load(**lora.model_dump(exclude={"weight"}))
+                lora_info = context.models.load(lora.lora)
                 assert isinstance(lora_info.model, LoRAModelRaw)
                 yield (lora_info.model, lora.weight)
                 del lora_info
@@ -138,18 +127,18 @@ class SDXLPromptInvocationBase:
     def run_clip_compel(
         self,
         context: InvocationContext,
-        clip_field: ClipField,
+        clip_field: CLIPField,
         prompt: str,
         get_pooled: bool,
         lora_prefix: str,
         zero_on_empty: bool,
     ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[ExtraConditioningInfo]]:
-        tokenizer_info = context.models.load(**clip_field.tokenizer.model_dump())
+        tokenizer_info = context.models.load(clip_field.tokenizer)
         tokenizer_model = tokenizer_info.model
         assert isinstance(tokenizer_model, CLIPTokenizer)
-        text_encoder_info = context.models.load(**clip_field.text_encoder.model_dump())
+        text_encoder_info = context.models.load(clip_field.text_encoder)
         text_encoder_model = text_encoder_info.model
-        assert isinstance(text_encoder_model, CLIPTextModel)
+        assert isinstance(text_encoder_model, (CLIPTextModel, CLIPTextModelWithProjection))
 
         # return zero on empty
         if prompt == "" and zero_on_empty:
@@ -174,7 +163,7 @@ class SDXLPromptInvocationBase:
 
         def _lora_loader() -> Iterator[Tuple[LoRAModelRaw, float]]:
             for lora in clip_field.loras:
-                lora_info = context.models.load(**lora.model_dump(exclude={"weight"}))
+                lora_info = context.models.load(lora.lora)
                 lora_model = lora_info.model
                 assert isinstance(lora_model, LoRAModelRaw)
                 yield (lora_model, lora.weight)
@@ -196,7 +185,8 @@ class SDXLPromptInvocationBase:
             # Apply CLIP Skip after LoRA to prevent LoRA application from failing on skipped layers.
             ModelPatcher.apply_clip_skip(text_encoder_model, clip_field.skipped_layers),
         ):
-            assert isinstance(text_encoder, CLIPTextModel)
+            assert isinstance(text_encoder, (CLIPTextModel, CLIPTextModelWithProjection))
+            text_encoder = cast(CLIPTextModel, text_encoder)
             compel = Compel(
                 tokenizer=tokenizer,
                 text_encoder=text_encoder,
@@ -263,8 +253,8 @@ class SDXLCompelPromptInvocation(BaseInvocation, SDXLPromptInvocationBase):
     crop_left: int = InputField(default=0, description="")
     target_width: int = InputField(default=1024, description="")
     target_height: int = InputField(default=1024, description="")
-    clip: ClipField = InputField(description=FieldDescriptions.clip, input=Input.Connection, title="CLIP 1")
-    clip2: ClipField = InputField(description=FieldDescriptions.clip, input=Input.Connection, title="CLIP 2")
+    clip: CLIPField = InputField(description=FieldDescriptions.clip, input=Input.Connection, title="CLIP 1")
+    clip2: CLIPField = InputField(description=FieldDescriptions.clip, input=Input.Connection, title="CLIP 2")
 
     @torch.no_grad()
     def invoke(self, context: InvocationContext) -> ConditioningOutput:
@@ -350,7 +340,7 @@ class SDXLRefinerCompelPromptInvocation(BaseInvocation, SDXLPromptInvocationBase
     crop_top: int = InputField(default=0, description="")
     crop_left: int = InputField(default=0, description="")
     aesthetic_score: float = InputField(default=6.0, description=FieldDescriptions.sdxl_aesthetic)
-    clip2: ClipField = InputField(description=FieldDescriptions.clip, input=Input.Connection)
+    clip2: CLIPField = InputField(description=FieldDescriptions.clip, input=Input.Connection)
 
     @torch.no_grad()
     def invoke(self, context: InvocationContext) -> ConditioningOutput:
@@ -380,10 +370,10 @@ class SDXLRefinerCompelPromptInvocation(BaseInvocation, SDXLPromptInvocationBase
 
 
 @invocation_output("clip_skip_output")
-class ClipSkipInvocationOutput(BaseInvocationOutput):
-    """Clip skip node output"""
+class CLIPSkipInvocationOutput(BaseInvocationOutput):
+    """CLIP skip node output"""
 
-    clip: Optional[ClipField] = OutputField(default=None, description=FieldDescriptions.clip, title="CLIP")
+    clip: Optional[CLIPField] = OutputField(default=None, description=FieldDescriptions.clip, title="CLIP")
 
 
 @invocation(
@@ -393,15 +383,15 @@ class ClipSkipInvocationOutput(BaseInvocationOutput):
     category="conditioning",
     version="1.0.0",
 )
-class ClipSkipInvocation(BaseInvocation):
+class CLIPSkipInvocation(BaseInvocation):
     """Skip layers in clip text_encoder model."""
 
-    clip: ClipField = InputField(description=FieldDescriptions.clip, input=Input.Connection, title="CLIP")
+    clip: CLIPField = InputField(description=FieldDescriptions.clip, input=Input.Connection, title="CLIP")
     skipped_layers: int = InputField(default=0, ge=0, description=FieldDescriptions.skipped_layers)
 
-    def invoke(self, context: InvocationContext) -> ClipSkipInvocationOutput:
+    def invoke(self, context: InvocationContext) -> CLIPSkipInvocationOutput:
         self.clip.skipped_layers += self.skipped_layers
-        return ClipSkipInvocationOutput(
+        return CLIPSkipInvocationOutput(
             clip=self.clip,
         )
 

invokeai/app/invocations/controlnet_image_processors.py

@@ -31,9 +31,11 @@ from invokeai.app.invocations.fields import (
     Input,
     InputField,
     OutputField,
+    UIType,
     WithBoard,
     WithMetadata,
 )
+from invokeai.app.invocations.model import ModelIdentifierField
 from invokeai.app.invocations.primitives import ImageOutput
 from invokeai.app.invocations.util import validate_begin_end_step, validate_weights
 from invokeai.app.services.shared.invocation_context import InvocationContext
@@ -51,15 +53,9 @@ CONTROLNET_RESIZE_VALUES = Literal[
 ]
 
 
-class ControlNetModelField(BaseModel):
-    """ControlNet model field"""
-
-    key: str = Field(description="Model config record key for the ControlNet model")
-
-
 class ControlField(BaseModel):
     image: ImageField = Field(description="The control image")
-    control_model: ControlNetModelField = Field(description="The ControlNet model to use")
+    control_model: ModelIdentifierField = Field(description="The ControlNet model to use")
     control_weight: Union[float, List[float]] = Field(default=1, description="The weight given to the ControlNet")
     begin_step_percent: float = Field(
         default=0, ge=0, le=1, description="When the ControlNet is first applied (% of total steps)"
@@ -95,7 +91,9 @@ class ControlNetInvocation(BaseInvocation):
     """Collects ControlNet info to pass to other nodes"""
 
     image: ImageField = InputField(description="The control image")
-    control_model: ControlNetModelField = InputField(description=FieldDescriptions.controlnet_model, input=Input.Direct)
+    control_model: ModelIdentifierField = InputField(
+        description=FieldDescriptions.controlnet_model, input=Input.Direct, ui_type=UIType.ControlNetModel
+    )
     control_weight: Union[float, List[float]] = InputField(
         default=1.0, ge=-1, le=2, description="The weight given to the ControlNet"
     )
@@ -576,7 +574,7 @@ DEPTH_ANYTHING_MODEL_SIZES = Literal["large", "base", "small"]
     title="Depth Anything Processor",
     tags=["controlnet", "depth", "depth anything"],
     category="controlnet",
-    version="1.0.0",
+    version="1.0.1",
 )
 class DepthAnythingImageProcessorInvocation(ImageProcessorInvocation):
     """Generates a depth map based on the Depth Anything algorithm"""
@@ -585,13 +583,12 @@ class DepthAnythingImageProcessorInvocation(ImageProcessorInvocation):
         default="small", description="The size of the depth model to use"
     )
     resolution: int = InputField(default=512, ge=64, multiple_of=64, description=FieldDescriptions.image_res)
-    offload: bool = InputField(default=False)
 
     def run_processor(self, image: Image.Image):
         depth_anything_detector = DepthAnythingDetector()
         depth_anything_detector.load_model(model_size=self.model_size)
 
-        processed_image = depth_anything_detector(image=image, resolution=self.resolution, offload=self.offload)
+        processed_image = depth_anything_detector(image=image, resolution=self.resolution)
         return processed_image
 
 

invokeai/app/invocations/fields.py

@@ -39,13 +39,15 @@ class UIType(str, Enum, metaclass=MetaEnum):
     """
 
     # region Model Field Types
+    MainModel = "MainModelField"
     SDXLMainModel = "SDXLMainModelField"
     SDXLRefinerModel = "SDXLRefinerModelField"
     ONNXModel = "ONNXModelField"
-    VaeModel = "VAEModelField"
+    VAEModel = "VAEModelField"
     LoRAModel = "LoRAModelField"
     ControlNetModel = "ControlNetModelField"
     IPAdapterModel = "IPAdapterModelField"
+    T2IAdapterModel = "T2IAdapterModelField"
     # endregion
 
     # region Misc Field Types
@@ -86,7 +88,6 @@ class UIType(str, Enum, metaclass=MetaEnum):
     IntegerPolymorphic = "DEPRECATED_IntegerPolymorphic"
     LatentsPolymorphic = "DEPRECATED_LatentsPolymorphic"
     StringPolymorphic = "DEPRECATED_StringPolymorphic"
-    MainModel = "DEPRECATED_MainModel"
     UNet = "DEPRECATED_UNet"
     Vae = "DEPRECATED_Vae"
     CLIP = "DEPRECATED_CLIP"
@@ -228,7 +229,7 @@ class ConditioningField(BaseModel):
     # endregion
 
 
-class MetadataField(RootModel):
+class MetadataField(RootModel[dict[str, Any]]):
     """
     Pydantic model for metadata with custom root of type dict[str, Any].
     Metadata is stored without a strict schema.

invokeai/app/invocations/ip_adapter.py

@@ -10,26 +10,18 @@ from invokeai.app.invocations.baseinvocation import (
     invocation,
     invocation_output,
 )
-from invokeai.app.invocations.fields import FieldDescriptions, Input, InputField, OutputField
+from invokeai.app.invocations.fields import FieldDescriptions, Input, InputField, OutputField, UIType
+from invokeai.app.invocations.model import ModelIdentifierField
 from invokeai.app.invocations.primitives import ImageField
 from invokeai.app.invocations.util import validate_begin_end_step, validate_weights
 from invokeai.app.services.shared.invocation_context import InvocationContext
-from invokeai.backend.model_manager.config import BaseModelType, ModelType
-
-
-# LS: Consider moving these two classes into model.py
-class IPAdapterModelField(BaseModel):
-    key: str = Field(description="Key to the IP-Adapter model")
-
-
-class CLIPVisionModelField(BaseModel):
-    key: str = Field(description="Key to the CLIP Vision image encoder model")
+from invokeai.backend.model_manager.config import AnyModelConfig, BaseModelType, IPAdapterConfig, ModelType
 
 
 class IPAdapterField(BaseModel):
     image: Union[ImageField, List[ImageField]] = Field(description="The IP-Adapter image prompt(s).")
-    ip_adapter_model: IPAdapterModelField = Field(description="The IP-Adapter model to use.")
-    image_encoder_model: CLIPVisionModelField = Field(description="The name of the CLIP image encoder model.")
+    ip_adapter_model: ModelIdentifierField = Field(description="The IP-Adapter model to use.")
+    image_encoder_model: ModelIdentifierField = Field(description="The name of the CLIP image encoder model.")
     weight: Union[float, List[float]] = Field(default=1, description="The weight given to the ControlNet")
     begin_step_percent: float = Field(
         default=0, ge=0, le=1, description="When the IP-Adapter is first applied (% of total steps)"
@@ -62,8 +54,12 @@ class IPAdapterInvocation(BaseInvocation):
 
     # Inputs
     image: Union[ImageField, List[ImageField]] = InputField(description="The IP-Adapter image prompt(s).")
-    ip_adapter_model: IPAdapterModelField = InputField(
-        description="The IP-Adapter model.", title="IP-Adapter Model", input=Input.Direct, ui_order=-1
+    ip_adapter_model: ModelIdentifierField = InputField(
+        description="The IP-Adapter model.",
+        title="IP-Adapter Model",
+        input=Input.Direct,
+        ui_order=-1,
+        ui_type=UIType.IPAdapterModel,
     )
 
     weight: Union[float, List[float]] = InputField(
@@ -90,20 +86,35 @@ class IPAdapterInvocation(BaseInvocation):
     def invoke(self, context: InvocationContext) -> IPAdapterOutput:
         # Lookup the CLIP Vision encoder that is intended to be used with the IP-Adapter model.
         ip_adapter_info = context.models.get_config(self.ip_adapter_model.key)
+        assert isinstance(ip_adapter_info, IPAdapterConfig)
         image_encoder_model_id = ip_adapter_info.image_encoder_model_id
         image_encoder_model_name = image_encoder_model_id.split("/")[-1].strip()
-        image_encoder_models = context.models.search_by_attrs(
-            model_name=image_encoder_model_name, base_model=BaseModelType.Any, model_type=ModelType.CLIPVision
-        )
-        assert len(image_encoder_models) == 1
-        image_encoder_model = CLIPVisionModelField(key=image_encoder_models[0].key)
+        image_encoder_model = self._get_image_encoder(context, image_encoder_model_name)
         return IPAdapterOutput(
             ip_adapter=IPAdapterField(
                 image=self.image,
                 ip_adapter_model=self.ip_adapter_model,
-                image_encoder_model=image_encoder_model,
+                image_encoder_model=ModelIdentifierField.from_config(image_encoder_model),
                 weight=self.weight,
                 begin_step_percent=self.begin_step_percent,
                 end_step_percent=self.end_step_percent,
             ),
         )
+
+    def _get_image_encoder(self, context: InvocationContext, image_encoder_model_name: str) -> AnyModelConfig:
+        found = False
+        while not found:
+            image_encoder_models = context.models.search_by_attrs(
+                name=image_encoder_model_name, base=BaseModelType.Any, type=ModelType.CLIPVision
+            )
+            found = len(image_encoder_models) > 0
+            if not found:
+                context.logger.warning(
+                    f"The image encoder required by this IP Adapter ({image_encoder_model_name}) is not installed."
+                )
+                context.logger.warning("Downloading and installing now. This may take a while.")
+                installer = context._services.model_manager.install
+                job = installer.heuristic_import(f"InvokeAI/{image_encoder_model_name}")
+                installer.wait_for_job(job, timeout=600)  # wait up to 10 minutes - then raise a TimeoutException
+        assert len(image_encoder_models) == 1
+        return image_encoder_models[0]

invokeai/app/invocations/latent.py

@@ -26,6 +26,7 @@ from diffusers.schedulers import SchedulerMixin as Scheduler
 from PIL import Image, ImageFilter
 from pydantic import field_validator
 from torchvision.transforms.functional import resize as tv_resize
+from transformers import CLIPVisionModelWithProjection
 
 from invokeai.app.invocations.constants import LATENT_SCALE_FACTOR, SCHEDULER_NAME_VALUES
 from invokeai.app.invocations.fields import (
@@ -65,7 +66,6 @@ from ...backend.stable_diffusion.diffusers_pipeline import (
     T2IAdapterData,
     image_resized_to_grid_as_tensor,
 )
-from ...backend.stable_diffusion.diffusion.shared_invokeai_diffusion import PostprocessingSettings
 from ...backend.stable_diffusion.schedulers import SCHEDULER_MAP
 from ...backend.util.devices import choose_precision, choose_torch_device
 from .baseinvocation import (
@@ -75,7 +75,7 @@ from .baseinvocation import (
     invocation_output,
 )
 from .controlnet_image_processors import ControlField
-from .model import ModelInfo, UNetField, VaeField
+from .model import ModelIdentifierField, UNetField, VAEField
 
 if choose_torch_device() == torch.device("mps"):
     from torch import mps
@@ -118,7 +118,7 @@ class SchedulerInvocation(BaseInvocation):
 class CreateDenoiseMaskInvocation(BaseInvocation):
     """Creates mask for denoising model run."""
 
-    vae: VaeField = InputField(description=FieldDescriptions.vae, input=Input.Connection, ui_order=0)
+    vae: VAEField = InputField(description=FieldDescriptions.vae, input=Input.Connection, ui_order=0)
     image: Optional[ImageField] = InputField(default=None, description="Image which will be masked", ui_order=1)
     mask: ImageField = InputField(description="The mask to use when pasting", ui_order=2)
     tiled: bool = InputField(default=False, description=FieldDescriptions.tiled, ui_order=3)
@@ -153,7 +153,7 @@ class CreateDenoiseMaskInvocation(BaseInvocation):
         )
 
         if image_tensor is not None:
-            vae_info = context.models.load(**self.vae.vae.model_dump())
+            vae_info = context.models.load(self.vae.vae)
 
             img_mask = tv_resize(mask, image_tensor.shape[-2:], T.InterpolationMode.BILINEAR, antialias=False)
             masked_image = image_tensor * torch.where(img_mask < 0.5, 0.0, 1.0)
@@ -173,6 +173,16 @@ class CreateDenoiseMaskInvocation(BaseInvocation):
         )
 
 
+@invocation_output("gradient_mask_output")
+class GradientMaskOutput(BaseInvocationOutput):
+    """Outputs a denoise mask and an image representing the total gradient of the mask."""
+
+    denoise_mask: DenoiseMaskField = OutputField(description="Mask for denoise model run")
+    expanded_mask_area: ImageField = OutputField(
+        description="Image representing the total gradient area of the mask. For paste-back purposes."
+    )
+
+
 @invocation(
     "create_gradient_mask",
     title="Create Gradient Mask",
@@ -193,49 +203,53 @@ class CreateGradientMaskInvocation(BaseInvocation):
     )
 
     @torch.no_grad()
-    def invoke(self, context: InvocationContext) -> DenoiseMaskOutput:
+    def invoke(self, context: InvocationContext) -> GradientMaskOutput:
         mask_image = context.images.get_pil(self.mask.image_name, mode="L")
+        if self.edge_radius > 0:
             if self.coherence_mode == "Box Blur":
                 blur_mask = mask_image.filter(ImageFilter.BoxBlur(self.edge_radius))
             else:  # Gaussian Blur OR Staged
                 # Gaussian Blur uses standard deviation. 1/2 radius is a good approximation
                 blur_mask = mask_image.filter(ImageFilter.GaussianBlur(self.edge_radius / 2))
 
-        mask_tensor: torch.Tensor = image_resized_to_grid_as_tensor(mask_image, normalize=False)
             blur_tensor: torch.Tensor = image_resized_to_grid_as_tensor(blur_mask, normalize=False)
 
-        # redistribute blur so that the edges are 0 and blur out to 1
+            # redistribute blur so that the original edges are 0 and blur outwards to 1
             blur_tensor = (blur_tensor - 0.5) * 2
 
             threshold = 1 - self.minimum_denoise
 
             if self.coherence_mode == "Staged":
-            # wherever the blur_tensor is masked to any degree, convert it to threshold
-            blur_tensor = torch.where((blur_tensor < 1), threshold, blur_tensor)
+                # wherever the blur_tensor is less than fully masked, convert it to threshold
+                blur_tensor = torch.where((blur_tensor < 1) & (blur_tensor > 0), threshold, blur_tensor)
             else:
                 # wherever the blur_tensor is above threshold but less than 1, drop it to threshold
                 blur_tensor = torch.where((blur_tensor > threshold) & (blur_tensor < 1), threshold, blur_tensor)
 
-        # multiply original mask to force actually masked regions to 0
-        blur_tensor = mask_tensor * blur_tensor
+        else:
+            blur_tensor: torch.Tensor = image_resized_to_grid_as_tensor(mask_image, normalize=False)
 
         mask_name = context.tensors.save(tensor=blur_tensor.unsqueeze(1))
 
-        return DenoiseMaskOutput.build(
-            mask_name=mask_name,
-            masked_latents_name=None,
-            gradient=True,
+        # compute a [0, 1] mask from the blur_tensor
+        expanded_mask = torch.where((blur_tensor < 1), 0, 1)
+        expanded_mask_image = Image.fromarray((expanded_mask.squeeze(0).numpy() * 255).astype(np.uint8), mode="L")
+        expanded_image_dto = context.images.save(expanded_mask_image)
+
+        return GradientMaskOutput(
+            denoise_mask=DenoiseMaskField(mask_name=mask_name, masked_latents_name=None, gradient=True),
+            expanded_mask_area=ImageField(image_name=expanded_image_dto.image_name),
         )
 
 
 def get_scheduler(
     context: InvocationContext,
-    scheduler_info: ModelInfo,
+    scheduler_info: ModelIdentifierField,
     scheduler_name: str,
     seed: int,
 ) -> Scheduler:
     scheduler_class, scheduler_extra_config = SCHEDULER_MAP.get(scheduler_name, SCHEDULER_MAP["ddim"])
-    orig_scheduler_info = context.models.load(**scheduler_info.model_dump())
+    orig_scheduler_info = context.models.load(scheduler_info)
     with orig_scheduler_info as orig_scheduler:
         scheduler_config = orig_scheduler.config
 
@@ -360,7 +374,6 @@ class DenoiseLatentsInvocation(BaseInvocation):
     ) -> ConditioningData:
         positive_cond_data = context.conditioning.load(self.positive_conditioning.conditioning_name)
         c = positive_cond_data.conditionings[0].to(device=unet.device, dtype=unet.dtype)
-        extra_conditioning_info = c.extra_conditioning
 
         negative_cond_data = context.conditioning.load(self.negative_conditioning.conditioning_name)
         uc = negative_cond_data.conditionings[0].to(device=unet.device, dtype=unet.dtype)
@@ -370,13 +383,6 @@ class DenoiseLatentsInvocation(BaseInvocation):
             text_embeddings=c,
             guidance_scale=self.cfg_scale,
             guidance_rescale_multiplier=self.cfg_rescale_multiplier,
-            extra=extra_conditioning_info,
-            postprocessing_settings=PostprocessingSettings(
-                threshold=0.0,  # threshold,
-                warmup=0.2,  # warmup,
-                h_symmetry_time_pct=None,  # h_symmetry_time_pct,
-                v_symmetry_time_pct=None,  # v_symmetry_time_pct,
-            ),
         )
 
         conditioning_data = conditioning_data.add_scheduler_args_if_applicable(  # FIXME
@@ -449,7 +455,7 @@ class DenoiseLatentsInvocation(BaseInvocation):
         #        and if weight is None, populate with default 1.0?
         controlnet_data = []
         for control_info in control_list:
-            control_model = exit_stack.enter_context(context.models.load(key=control_info.control_model.key))
+            control_model = exit_stack.enter_context(context.models.load(control_info.control_model))
 
             # control_models.append(control_model)
             control_image_field = control_info.image
@@ -511,11 +517,10 @@ class DenoiseLatentsInvocation(BaseInvocation):
         conditioning_data.ip_adapter_conditioning = []
         for single_ip_adapter in ip_adapter:
             ip_adapter_model: Union[IPAdapter, IPAdapterPlus] = exit_stack.enter_context(
-                context.models.load(key=single_ip_adapter.ip_adapter_model.key)
+                context.models.load(single_ip_adapter.ip_adapter_model)
             )
 
-            image_encoder_model_info = context.models.load(key=single_ip_adapter.image_encoder_model.key)
-
+            image_encoder_model_info = context.models.load(single_ip_adapter.image_encoder_model)
             # `single_ip_adapter.image` could be a list or a single ImageField. Normalize to a list here.
             single_ipa_image_fields = single_ip_adapter.image
             if not isinstance(single_ipa_image_fields, list):
@@ -526,6 +531,7 @@ class DenoiseLatentsInvocation(BaseInvocation):
             # TODO(ryand): With some effort, the step of running the CLIP Vision encoder could be done before any other
             # models are needed in memory. This would help to reduce peak memory utilization in low-memory environments.
             with image_encoder_model_info as image_encoder_model:
+                assert isinstance(image_encoder_model, CLIPVisionModelWithProjection)
                 # Get image embeddings from CLIP and ImageProjModel.
                 image_prompt_embeds, uncond_image_prompt_embeds = ip_adapter_model.get_image_embeds(
                     single_ipa_images, image_encoder_model
@@ -565,8 +571,8 @@ class DenoiseLatentsInvocation(BaseInvocation):
 
         t2i_adapter_data = []
         for t2i_adapter_field in t2i_adapter:
-            t2i_adapter_model_config = context.models.get_config(key=t2i_adapter_field.t2i_adapter_model.key)
-            t2i_adapter_loaded_model = context.models.load(key=t2i_adapter_field.t2i_adapter_model.key)
+            t2i_adapter_model_config = context.models.get_config(t2i_adapter_field.t2i_adapter_model.key)
+            t2i_adapter_loaded_model = context.models.load(t2i_adapter_field.t2i_adapter_model)
             image = context.images.get_pil(t2i_adapter_field.image.image_name)
 
             # The max_unet_downscale is the maximum amount that the UNet model downscales the latent image internally.
@@ -671,7 +677,7 @@ class DenoiseLatentsInvocation(BaseInvocation):
         if self.denoise_mask.masked_latents_name is not None:
             masked_latents = context.tensors.load(self.denoise_mask.masked_latents_name)
         else:
-            masked_latents = None
+            masked_latents = torch.where(mask < 0.5, 0.0, latents)
 
         return 1 - mask, masked_latents, self.denoise_mask.gradient
 
@@ -719,12 +725,13 @@ class DenoiseLatentsInvocation(BaseInvocation):
 
             def _lora_loader() -> Iterator[Tuple[LoRAModelRaw, float]]:
                 for lora in self.unet.loras:
-                    lora_info = context.models.load(**lora.model_dump(exclude={"weight"}))
+                    lora_info = context.models.load(lora.lora)
+                    assert isinstance(lora_info.model, LoRAModelRaw)
                     yield (lora_info.model, lora.weight)
                     del lora_info
                 return
 
-            unet_info = context.models.load(**self.unet.unet.model_dump())
+            unet_info = context.models.load(self.unet.unet)
             assert isinstance(unet_info.model, UNet2DConditionModel)
             with (
                 ExitStack() as exit_stack,
@@ -777,10 +784,7 @@ class DenoiseLatentsInvocation(BaseInvocation):
                     denoising_end=self.denoising_end,
                 )
 
-                (
-                    result_latents,
-                    result_attention_map_saver,
-                ) = pipeline.latents_from_embeddings(
+                result_latents = pipeline.latents_from_embeddings(
                     latents=latents,
                     timesteps=timesteps,
                     init_timestep=init_timestep,
@@ -821,7 +825,7 @@ class LatentsToImageInvocation(BaseInvocation, WithMetadata, WithBoard):
         description=FieldDescriptions.latents,
         input=Input.Connection,
     )
-    vae: VaeField = InputField(
+    vae: VAEField = InputField(
         description=FieldDescriptions.vae,
         input=Input.Connection,
     )
@@ -832,15 +836,15 @@ class LatentsToImageInvocation(BaseInvocation, WithMetadata, WithBoard):
     def invoke(self, context: InvocationContext) -> ImageOutput:
         latents = context.tensors.load(self.latents.latents_name)
 
-        vae_info = context.models.load(**self.vae.vae.model_dump())
-
+        vae_info = context.models.load(self.vae.vae)
+        assert isinstance(vae_info.model, (UNet2DConditionModel, AutoencoderKL, AutoencoderTiny))
         with set_seamless(vae_info.model, self.vae.seamless_axes), vae_info as vae:
             assert isinstance(vae, torch.nn.Module)
             latents = latents.to(vae.device)
             if self.fp32:
                 vae.to(dtype=torch.float32)
 
-                use_torch_2_0_or_xformers = isinstance(
+                use_torch_2_0_or_xformers = hasattr(vae.decoder, "mid_block") and isinstance(
                     vae.decoder.mid_block.attentions[0].processor,
                     (
                         AttnProcessor2_0,
@@ -999,7 +1003,7 @@ class ImageToLatentsInvocation(BaseInvocation):
     image: ImageField = InputField(
         description="The image to encode",
     )
-    vae: VaeField = InputField(
+    vae: VAEField = InputField(
         description=FieldDescriptions.vae,
         input=Input.Connection,
     )
@@ -1014,7 +1018,7 @@ class ImageToLatentsInvocation(BaseInvocation):
             if upcast:
                 vae.to(dtype=torch.float32)
 
-                use_torch_2_0_or_xformers = isinstance(
+                use_torch_2_0_or_xformers = hasattr(vae.decoder, "mid_block") and isinstance(
                     vae.decoder.mid_block.attentions[0].processor,
                     (
                         AttnProcessor2_0,
@@ -1055,7 +1059,7 @@ class ImageToLatentsInvocation(BaseInvocation):
     def invoke(self, context: InvocationContext) -> LatentsOutput:
         image = context.images.get_pil(self.image.image_name)
 
-        vae_info = context.models.load(**self.vae.vae.model_dump())
+        vae_info = context.models.load(self.vae.vae)
 
         image_tensor = image_resized_to_grid_as_tensor(image.convert("RGB"))
         if image_tensor.dim() == 3:

invokeai/app/invocations/metadata.py

@@ -8,7 +8,10 @@ from invokeai.app.invocations.baseinvocation import (
     invocation,
     invocation_output,
 )
-from invokeai.app.invocations.controlnet_image_processors import ControlField
+from invokeai.app.invocations.controlnet_image_processors import (
+    CONTROLNET_MODE_VALUES,
+    CONTROLNET_RESIZE_VALUES,
+)
 from invokeai.app.invocations.fields import (
     FieldDescriptions,
     ImageField,
@@ -17,9 +20,7 @@ from invokeai.app.invocations.fields import (
     OutputField,
     UIType,
 )
-from invokeai.app.invocations.ip_adapter import IPAdapterModelField
-from invokeai.app.invocations.model import LoRAModelField, MainModelField, VAEModelField
-from invokeai.app.invocations.t2i_adapter import T2IAdapterField
+from invokeai.app.invocations.model import ModelIdentifierField
 from invokeai.app.services.shared.invocation_context import InvocationContext
 
 from ...version import __version__
@@ -33,7 +34,7 @@ class MetadataItemField(BaseModel):
 class LoRAMetadataField(BaseModel):
     """LoRA Metadata Field"""
 
-    model: LoRAModelField = Field(description=FieldDescriptions.lora_model)
+    model: ModelIdentifierField = Field(description=FieldDescriptions.lora_model)
     weight: float = Field(description=FieldDescriptions.lora_weight)
 
 
@@ -41,16 +42,41 @@ class IPAdapterMetadataField(BaseModel):
     """IP Adapter Field, minus the CLIP Vision Encoder model"""
 
     image: ImageField = Field(description="The IP-Adapter image prompt.")
-    ip_adapter_model: IPAdapterModelField = Field(
-        description="The IP-Adapter model.",
-    )
-    weight: Union[float, list[float]] = Field(
-        description="The weight given to the IP-Adapter",
-    )
+    ip_adapter_model: ModelIdentifierField = Field(description="The IP-Adapter model.")
+    weight: Union[float, list[float]] = Field(description="The weight given to the IP-Adapter")
     begin_step_percent: float = Field(description="When the IP-Adapter is first applied (% of total steps)")
     end_step_percent: float = Field(description="When the IP-Adapter is last applied (% of total steps)")
 
 
+class T2IAdapterMetadataField(BaseModel):
+    image: ImageField = Field(description="The control image.")
+    processed_image: Optional[ImageField] = Field(default=None, description="The control image, after processing.")
+    t2i_adapter_model: ModelIdentifierField = Field(description="The T2I-Adapter model to use.")
+    weight: Union[float, list[float]] = Field(default=1, description="The weight given to the T2I-Adapter")
+    begin_step_percent: float = Field(
+        default=0, ge=0, le=1, description="When the T2I-Adapter is first applied (% of total steps)"
+    )
+    end_step_percent: float = Field(
+        default=1, ge=0, le=1, description="When the T2I-Adapter is last applied (% of total steps)"
+    )
+    resize_mode: CONTROLNET_RESIZE_VALUES = Field(default="just_resize", description="The resize mode to use")
+
+
+class ControlNetMetadataField(BaseModel):
+    image: ImageField = Field(description="The control image")
+    processed_image: Optional[ImageField] = Field(default=None, description="The control image, after processing.")
+    control_model: ModelIdentifierField = Field(description="The ControlNet model to use")
+    control_weight: Union[float, list[float]] = Field(default=1, description="The weight given to the ControlNet")
+    begin_step_percent: float = Field(
+        default=0, ge=0, le=1, description="When the ControlNet is first applied (% of total steps)"
+    )
+    end_step_percent: float = Field(
+        default=1, ge=0, le=1, description="When the ControlNet is last applied (% of total steps)"
+    )
+    control_mode: CONTROLNET_MODE_VALUES = Field(default="balanced", description="The control mode to use")
+    resize_mode: CONTROLNET_RESIZE_VALUES = Field(default="just_resize", description="The resize mode to use")
+
+
 @invocation_output("metadata_item_output")
 class MetadataItemOutput(BaseInvocationOutput):
     """Metadata Item Output"""
@@ -140,14 +166,14 @@ class CoreMetadataInvocation(BaseInvocation):
         default=None,
         description="The number of skipped CLIP layers",
     )
-    model: Optional[MainModelField] = InputField(default=None, description="The main model used for inference")
-    controlnets: Optional[list[ControlField]] = InputField(
+    model: Optional[ModelIdentifierField] = InputField(default=None, description="The main model used for inference")
+    controlnets: Optional[list[ControlNetMetadataField]] = InputField(
         default=None, description="The ControlNets used for inference"
     )
     ipAdapters: Optional[list[IPAdapterMetadataField]] = InputField(
         default=None, description="The IP Adapters used for inference"
     )
-    t2iAdapters: Optional[list[T2IAdapterField]] = InputField(
+    t2iAdapters: Optional[list[T2IAdapterMetadataField]] = InputField(
         default=None, description="The IP Adapters used for inference"
     )
     loras: Optional[list[LoRAMetadataField]] = InputField(default=None, description="The LoRAs used for inference")
@@ -159,7 +185,7 @@ class CoreMetadataInvocation(BaseInvocation):
         default=None,
         description="The name of the initial image",
     )
-    vae: Optional[VAEModelField] = InputField(
+    vae: Optional[ModelIdentifierField] = InputField(
         default=None,
         description="The VAE used for decoding, if the main model's default was not used",
     )
@@ -190,7 +216,7 @@ class CoreMetadataInvocation(BaseInvocation):
     )
 
     # SDXL Refiner
-    refiner_model: Optional[MainModelField] = InputField(
+    refiner_model: Optional[ModelIdentifierField] = InputField(
         default=None,
         description="The SDXL Refiner model used",
     )
@@ -222,10 +248,9 @@ class CoreMetadataInvocation(BaseInvocation):
     def invoke(self, context: InvocationContext) -> MetadataOutput:
         """Collects and outputs a CoreMetadata object"""
 
-        return MetadataOutput(
-            metadata=MetadataField.model_validate(
-                self.model_dump(exclude_none=True, exclude={"id", "type", "is_intermediate", "use_cache"})
-            )
-        )
+        as_dict = self.model_dump(exclude_none=True, exclude={"id", "type", "is_intermediate", "use_cache"})
+        as_dict["app_version"] = __version__
+
+        return MetadataOutput(metadata=MetadataField.model_validate(as_dict))
 
     model_config = ConfigDict(extra="allow")

invokeai/app/invocations/model.py

@@ -3,11 +3,11 @@ from typing import List, Optional
 
 from pydantic import BaseModel, Field
 
-from invokeai.app.invocations.fields import FieldDescriptions, Input, InputField, OutputField
+from invokeai.app.invocations.fields import FieldDescriptions, Input, InputField, OutputField, UIType
 from invokeai.app.services.shared.invocation_context import InvocationContext
 from invokeai.app.shared.models import FreeUConfig
+from invokeai.backend.model_manager.config import AnyModelConfig, BaseModelType, ModelType, SubModelType
 
-from ...backend.model_manager import SubModelType
 from .baseinvocation import (
     BaseInvocation,
     BaseInvocationOutput,
@@ -16,33 +16,52 @@ from .baseinvocation import (
 )
 
 
-class ModelInfo(BaseModel):
-    key: str = Field(description="Key of model as returned by ModelRecordServiceBase.get_model()")
-    submodel_type: Optional[SubModelType] = Field(default=None, description="Info to load submodel")
+class ModelIdentifierField(BaseModel):
+    key: str = Field(description="The model's unique key")
+    hash: str = Field(description="The model's BLAKE3 hash")
+    name: str = Field(description="The model's name")
+    base: BaseModelType = Field(description="The model's base model type")
+    type: ModelType = Field(description="The model's type")
+    submodel_type: Optional[SubModelType] = Field(
+        description="The submodel to load, if this is a main model", default=None
+    )
+
+    @classmethod
+    def from_config(
+        cls, config: "AnyModelConfig", submodel_type: Optional[SubModelType] = None
+    ) -> "ModelIdentifierField":
+        return cls(
+            key=config.key,
+            hash=config.hash,
+            name=config.name,
+            base=config.base,
+            type=config.type,
+            submodel_type=submodel_type,
+        )
 
 
-class LoraInfo(ModelInfo):
-    weight: float = Field(description="Lora's weight which to use when apply to model")
+class LoRAField(BaseModel):
+    lora: ModelIdentifierField = Field(description="Info to load lora model")
+    weight: float = Field(description="Weight to apply to lora model")
 
 
 class UNetField(BaseModel):
-    unet: ModelInfo = Field(description="Info to load unet submodel")
-    scheduler: ModelInfo = Field(description="Info to load scheduler submodel")
-    loras: List[LoraInfo] = Field(description="Loras to apply on model loading")
+    unet: ModelIdentifierField = Field(description="Info to load unet submodel")
+    scheduler: ModelIdentifierField = Field(description="Info to load scheduler submodel")
+    loras: List[LoRAField] = Field(description="LoRAs to apply on model loading")
     seamless_axes: List[str] = Field(default_factory=list, description='Axes("x" and "y") to which apply seamless')
     freeu_config: Optional[FreeUConfig] = Field(default=None, description="FreeU configuration")
 
 
-class ClipField(BaseModel):
-    tokenizer: ModelInfo = Field(description="Info to load tokenizer submodel")
-    text_encoder: ModelInfo = Field(description="Info to load text_encoder submodel")
+class CLIPField(BaseModel):
+    tokenizer: ModelIdentifierField = Field(description="Info to load tokenizer submodel")
+    text_encoder: ModelIdentifierField = Field(description="Info to load text_encoder submodel")
     skipped_layers: int = Field(description="Number of skipped layers in text_encoder")
-    loras: List[LoraInfo] = Field(description="Loras to apply on model loading")
+    loras: List[LoRAField] = Field(description="LoRAs to apply on model loading")
 
 
-class VaeField(BaseModel):
-    # TODO: better naming?
-    vae: ModelInfo = Field(description="Info to load vae submodel")
+class VAEField(BaseModel):
+    vae: ModelIdentifierField = Field(description="Info to load vae submodel")
     seamless_axes: List[str] = Field(default_factory=list, description='Axes("x" and "y") to which apply seamless')
 
 
@@ -57,14 +76,14 @@ class UNetOutput(BaseInvocationOutput):
 class VAEOutput(BaseInvocationOutput):
     """Base class for invocations that output a VAE field"""
 
-    vae: VaeField = OutputField(description=FieldDescriptions.vae, title="VAE")
+    vae: VAEField = OutputField(description=FieldDescriptions.vae, title="VAE")
 
 
 @invocation_output("clip_output")
 class CLIPOutput(BaseInvocationOutput):
     """Base class for invocations that output a CLIP field"""
 
-    clip: ClipField = OutputField(description=FieldDescriptions.clip, title="CLIP")
+    clip: CLIPField = OutputField(description=FieldDescriptions.clip, title="CLIP")
 
 
 @invocation_output("model_loader_output")
@@ -74,18 +93,6 @@ class ModelLoaderOutput(UNetOutput, CLIPOutput, VAEOutput):
     pass
 
 
-class MainModelField(BaseModel):
-    """Main model field"""
-
-    key: str = Field(description="Model key")
-
-
-class LoRAModelField(BaseModel):
-    """LoRA model field"""
-
-    key: str = Field(description="LoRA model key")
-
-
 @invocation(
     "main_model_loader",
     title="Main Model",
@@ -96,62 +103,44 @@ class LoRAModelField(BaseModel):
 class MainModelLoaderInvocation(BaseInvocation):
     """Loads a main model, outputting its submodels."""
 
-    model: MainModelField = InputField(description=FieldDescriptions.main_model, input=Input.Direct)
+    model: ModelIdentifierField = InputField(
+        description=FieldDescriptions.main_model, input=Input.Direct, ui_type=UIType.MainModel
+    )
     # TODO: precision?
 
     def invoke(self, context: InvocationContext) -> ModelLoaderOutput:
-        key = self.model.key
-
         # TODO: not found exceptions
-        if not context.models.exists(key):
-            raise Exception(f"Unknown model {key}")
+        if not context.models.exists(self.model.key):
+            raise Exception(f"Unknown model {self.model.key}")
+
+        unet = self.model.model_copy(update={"submodel_type": SubModelType.UNet})
+        scheduler = self.model.model_copy(update={"submodel_type": SubModelType.Scheduler})
+        tokenizer = self.model.model_copy(update={"submodel_type": SubModelType.Tokenizer})
+        text_encoder = self.model.model_copy(update={"submodel_type": SubModelType.TextEncoder})
+        vae = self.model.model_copy(update={"submodel_type": SubModelType.VAE})
 
         return ModelLoaderOutput(
-            unet=UNetField(
-                unet=ModelInfo(
-                    key=key,
-                    submodel_type=SubModelType.UNet,
-                ),
-                scheduler=ModelInfo(
-                    key=key,
-                    submodel_type=SubModelType.Scheduler,
-                ),
-                loras=[],
-            ),
-            clip=ClipField(
-                tokenizer=ModelInfo(
-                    key=key,
-                    submodel_type=SubModelType.Tokenizer,
-                ),
-                text_encoder=ModelInfo(
-                    key=key,
-                    submodel_type=SubModelType.TextEncoder,
-                ),
-                loras=[],
-                skipped_layers=0,
-            ),
-            vae=VaeField(
-                vae=ModelInfo(
-                    key=key,
-                    submodel_type=SubModelType.Vae,
-                ),
-            ),
+            unet=UNetField(unet=unet, scheduler=scheduler, loras=[]),
+            clip=CLIPField(tokenizer=tokenizer, text_encoder=text_encoder, loras=[], skipped_layers=0),
+            vae=VAEField(vae=vae),
         )
 
 
 @invocation_output("lora_loader_output")
-class LoraLoaderOutput(BaseInvocationOutput):
+class LoRALoaderOutput(BaseInvocationOutput):
     """Model loader output"""
 
     unet: Optional[UNetField] = OutputField(default=None, description=FieldDescriptions.unet, title="UNet")
-    clip: Optional[ClipField] = OutputField(default=None, description=FieldDescriptions.clip, title="CLIP")
+    clip: Optional[CLIPField] = OutputField(default=None, description=FieldDescriptions.clip, title="CLIP")
 
 
 @invocation("lora_loader", title="LoRA", tags=["model"], category="model", version="1.0.1")
-class LoraLoaderInvocation(BaseInvocation):
+class LoRALoaderInvocation(BaseInvocation):
     """Apply selected lora to unet and text_encoder."""
 
-    lora: LoRAModelField = InputField(description=FieldDescriptions.lora_model, input=Input.Direct, title="LoRA")
+    lora: ModelIdentifierField = InputField(
+        description=FieldDescriptions.lora_model, input=Input.Direct, title="LoRA", ui_type=UIType.LoRAModel
+    )
     weight: float = InputField(default=0.75, description=FieldDescriptions.lora_weight)
     unet: Optional[UNetField] = InputField(
         default=None,
@@ -159,46 +148,41 @@ class LoraLoaderInvocation(BaseInvocation):
         input=Input.Connection,
         title="UNet",
     )
-    clip: Optional[ClipField] = InputField(
+    clip: Optional[CLIPField] = InputField(
         default=None,
         description=FieldDescriptions.clip,
         input=Input.Connection,
         title="CLIP",
     )
 
-    def invoke(self, context: InvocationContext) -> LoraLoaderOutput:
-        if self.lora is None:
-            raise Exception("No LoRA provided")
-
+    def invoke(self, context: InvocationContext) -> LoRALoaderOutput:
         lora_key = self.lora.key
 
         if not context.models.exists(lora_key):
             raise Exception(f"Unkown lora: {lora_key}!")
 
-        if self.unet is not None and any(lora.key == lora_key for lora in self.unet.loras):
-            raise Exception(f'Lora "{lora_key}" already applied to unet')
+        if self.unet is not None and any(lora.lora.key == lora_key for lora in self.unet.loras):
+            raise Exception(f'LoRA "{lora_key}" already applied to unet')
 
-        if self.clip is not None and any(lora.key == lora_key for lora in self.clip.loras):
-            raise Exception(f'Lora "{lora_key}" already applied to clip')
+        if self.clip is not None and any(lora.lora.key == lora_key for lora in self.clip.loras):
+            raise Exception(f'LoRA "{lora_key}" already applied to clip')
 
-        output = LoraLoaderOutput()
+        output = LoRALoaderOutput()
 
         if self.unet is not None:
-            output.unet = copy.deepcopy(self.unet)
+            output.unet = self.unet.model_copy(deep=True)
             output.unet.loras.append(
-                LoraInfo(
-                    key=lora_key,
-                    submodel_type=None,
+                LoRAField(
+                    lora=self.lora,
                     weight=self.weight,
                 )
             )
 
         if self.clip is not None:
-            output.clip = copy.deepcopy(self.clip)
+            output.clip = self.clip.model_copy(deep=True)
             output.clip.loras.append(
-                LoraInfo(
-                    key=lora_key,
-                    submodel_type=None,
+                LoRAField(
+                    lora=self.lora,
                     weight=self.weight,
                 )
             )
@@ -207,12 +191,12 @@ class LoraLoaderInvocation(BaseInvocation):
 
 
 @invocation_output("sdxl_lora_loader_output")
-class SDXLLoraLoaderOutput(BaseInvocationOutput):
+class SDXLLoRALoaderOutput(BaseInvocationOutput):
     """SDXL LoRA Loader Output"""
 
     unet: Optional[UNetField] = OutputField(default=None, description=FieldDescriptions.unet, title="UNet")
-    clip: Optional[ClipField] = OutputField(default=None, description=FieldDescriptions.clip, title="CLIP 1")
-    clip2: Optional[ClipField] = OutputField(default=None, description=FieldDescriptions.clip, title="CLIP 2")
+    clip: Optional[CLIPField] = OutputField(default=None, description=FieldDescriptions.clip, title="CLIP 1")
+    clip2: Optional[CLIPField] = OutputField(default=None, description=FieldDescriptions.clip, title="CLIP 2")
 
 
 @invocation(
@@ -222,10 +206,12 @@ class SDXLLoraLoaderOutput(BaseInvocationOutput):
     category="model",
     version="1.0.1",
 )
-class SDXLLoraLoaderInvocation(BaseInvocation):
+class SDXLLoRALoaderInvocation(BaseInvocation):
     """Apply selected lora to unet and text_encoder."""
 
-    lora: LoRAModelField = InputField(description=FieldDescriptions.lora_model, input=Input.Direct, title="LoRA")
+    lora: ModelIdentifierField = InputField(
+        description=FieldDescriptions.lora_model, input=Input.Direct, title="LoRA", ui_type=UIType.LoRAModel
+    )
     weight: float = InputField(default=0.75, description=FieldDescriptions.lora_weight)
     unet: Optional[UNetField] = InputField(
         default=None,
@@ -233,65 +219,59 @@ class SDXLLoraLoaderInvocation(BaseInvocation):
         input=Input.Connection,
         title="UNet",
     )
-    clip: Optional[ClipField] = InputField(
+    clip: Optional[CLIPField] = InputField(
         default=None,
         description=FieldDescriptions.clip,
         input=Input.Connection,
         title="CLIP 1",
     )
-    clip2: Optional[ClipField] = InputField(
+    clip2: Optional[CLIPField] = InputField(
         default=None,
         description=FieldDescriptions.clip,
         input=Input.Connection,
         title="CLIP 2",
     )
 
-    def invoke(self, context: InvocationContext) -> SDXLLoraLoaderOutput:
-        if self.lora is None:
-            raise Exception("No LoRA provided")
-
+    def invoke(self, context: InvocationContext) -> SDXLLoRALoaderOutput:
         lora_key = self.lora.key
 
         if not context.models.exists(lora_key):
             raise Exception(f"Unknown lora: {lora_key}!")
 
-        if self.unet is not None and any(lora.key == lora_key for lora in self.unet.loras):
-            raise Exception(f'Lora "{lora_key}" already applied to unet')
+        if self.unet is not None and any(lora.lora.key == lora_key for lora in self.unet.loras):
+            raise Exception(f'LoRA "{lora_key}" already applied to unet')
 
-        if self.clip is not None and any(lora.key == lora_key for lora in self.clip.loras):
-            raise Exception(f'Lora "{lora_key}" already applied to clip')
+        if self.clip is not None and any(lora.lora.key == lora_key for lora in self.clip.loras):
+            raise Exception(f'LoRA "{lora_key}" already applied to clip')
 
-        if self.clip2 is not None and any(lora.key == lora_key for lora in self.clip2.loras):
-            raise Exception(f'Lora "{lora_key}" already applied to clip2')
+        if self.clip2 is not None and any(lora.lora.key == lora_key for lora in self.clip2.loras):
+            raise Exception(f'LoRA "{lora_key}" already applied to clip2')
 
-        output = SDXLLoraLoaderOutput()
+        output = SDXLLoRALoaderOutput()
 
         if self.unet is not None:
-            output.unet = copy.deepcopy(self.unet)
+            output.unet = self.unet.model_copy(deep=True)
             output.unet.loras.append(
-                LoraInfo(
-                    key=lora_key,
-                    submodel_type=None,
+                LoRAField(
+                    lora=self.lora,
                     weight=self.weight,
                 )
             )
 
         if self.clip is not None:
-            output.clip = copy.deepcopy(self.clip)
+            output.clip = self.clip.model_copy(deep=True)
             output.clip.loras.append(
-                LoraInfo(
-                    key=lora_key,
-                    submodel_type=None,
+                LoRAField(
+                    lora=self.lora,
                     weight=self.weight,
                 )
             )
 
         if self.clip2 is not None:
-            output.clip2 = copy.deepcopy(self.clip2)
+            output.clip2 = self.clip2.model_copy(deep=True)
             output.clip2.loras.append(
-                LoraInfo(
-                    key=lora_key,
-                    submodel_type=None,
+                LoRAField(
+                    lora=self.lora,
                     weight=self.weight,
                 )
             )
@@ -299,20 +279,12 @@ class SDXLLoraLoaderInvocation(BaseInvocation):
         return output
 
 
-class VAEModelField(BaseModel):
-    """Vae model field"""
-
-    key: str = Field(description="Model's key")
-
-
 @invocation("vae_loader", title="VAE", tags=["vae", "model"], category="model", version="1.0.1")
-class VaeLoaderInvocation(BaseInvocation):
+class VAELoaderInvocation(BaseInvocation):
     """Loads a VAE model, outputting a VaeLoaderOutput"""
 
-    vae_model: VAEModelField = InputField(
-        description=FieldDescriptions.vae_model,
-        input=Input.Direct,
-        title="VAE",
+    vae_model: ModelIdentifierField = InputField(
+        description=FieldDescriptions.vae_model, input=Input.Direct, title="VAE", ui_type=UIType.VAEModel
     )
 
     def invoke(self, context: InvocationContext) -> VAEOutput:
@@ -321,7 +293,7 @@ class VaeLoaderInvocation(BaseInvocation):
         if not context.models.exists(key):
             raise Exception(f"Unkown vae: {key}!")
 
-        return VAEOutput(vae=VaeField(vae=ModelInfo(key=key)))
+        return VAEOutput(vae=VAEField(vae=self.vae_model))
 
 
 @invocation_output("seamless_output")
@@ -329,7 +301,7 @@ class SeamlessModeOutput(BaseInvocationOutput):
     """Modified Seamless Model output"""
 
     unet: Optional[UNetField] = OutputField(default=None, description=FieldDescriptions.unet, title="UNet")
-    vae: Optional[VaeField] = OutputField(default=None, description=FieldDescriptions.vae, title="VAE")
+    vae: Optional[VAEField] = OutputField(default=None, description=FieldDescriptions.vae, title="VAE")
 
 
 @invocation(
@@ -348,7 +320,7 @@ class SeamlessModeInvocation(BaseInvocation):
         input=Input.Connection,
         title="UNet",
     )
-    vae: Optional[VaeField] = InputField(
+    vae: Optional[VAEField] = InputField(
         default=None,
         description=FieldDescriptions.vae_model,
         input=Input.Connection,

invokeai/app/invocations/sdxl.py

@@ -8,7 +8,7 @@ from .baseinvocation import (
     invocation,
     invocation_output,
 )
-from .model import ClipField, MainModelField, ModelInfo, UNetField, VaeField
+from .model import CLIPField, ModelIdentifierField, UNetField, VAEField
 
 
 @invocation_output("sdxl_model_loader_output")
@@ -16,9 +16,9 @@ class SDXLModelLoaderOutput(BaseInvocationOutput):
     """SDXL base model loader output"""
 
     unet: UNetField = OutputField(description=FieldDescriptions.unet, title="UNet")
-    clip: ClipField = OutputField(description=FieldDescriptions.clip, title="CLIP 1")
-    clip2: ClipField = OutputField(description=FieldDescriptions.clip, title="CLIP 2")
-    vae: VaeField = OutputField(description=FieldDescriptions.vae, title="VAE")
+    clip: CLIPField = OutputField(description=FieldDescriptions.clip, title="CLIP 1")
+    clip2: CLIPField = OutputField(description=FieldDescriptions.clip, title="CLIP 2")
+    vae: VAEField = OutputField(description=FieldDescriptions.vae, title="VAE")
 
 
 @invocation_output("sdxl_refiner_model_loader_output")
@@ -26,15 +26,15 @@ class SDXLRefinerModelLoaderOutput(BaseInvocationOutput):
     """SDXL refiner model loader output"""
 
     unet: UNetField = OutputField(description=FieldDescriptions.unet, title="UNet")
-    clip2: ClipField = OutputField(description=FieldDescriptions.clip, title="CLIP 2")
-    vae: VaeField = OutputField(description=FieldDescriptions.vae, title="VAE")
+    clip2: CLIPField = OutputField(description=FieldDescriptions.clip, title="CLIP 2")
+    vae: VAEField = OutputField(description=FieldDescriptions.vae, title="VAE")
 
 
 @invocation("sdxl_model_loader", title="SDXL Main Model", tags=["model", "sdxl"], category="model", version="1.0.1")
 class SDXLModelLoaderInvocation(BaseInvocation):
     """Loads an sdxl base model, outputting its submodels."""
 
-    model: MainModelField = InputField(
+    model: ModelIdentifierField = InputField(
         description=FieldDescriptions.sdxl_main_model, input=Input.Direct, ui_type=UIType.SDXLMainModel
     )
     # TODO: precision?
@@ -46,48 +46,19 @@ class SDXLModelLoaderInvocation(BaseInvocation):
         if not context.models.exists(model_key):
             raise Exception(f"Unknown model: {model_key}")
 
+        unet = self.model.model_copy(update={"submodel_type": SubModelType.UNet})
+        scheduler = self.model.model_copy(update={"submodel_type": SubModelType.Scheduler})
+        tokenizer = self.model.model_copy(update={"submodel_type": SubModelType.Tokenizer})
+        text_encoder = self.model.model_copy(update={"submodel_type": SubModelType.TextEncoder})
+        tokenizer2 = self.model.model_copy(update={"submodel_type": SubModelType.Tokenizer2})
+        text_encoder2 = self.model.model_copy(update={"submodel_type": SubModelType.TextEncoder2})
+        vae = self.model.model_copy(update={"submodel_type": SubModelType.VAE})
+
         return SDXLModelLoaderOutput(
-            unet=UNetField(
-                unet=ModelInfo(
-                    key=model_key,
-                    submodel_type=SubModelType.UNet,
-                ),
-                scheduler=ModelInfo(
-                    key=model_key,
-                    submodel_type=SubModelType.Scheduler,
-                ),
-                loras=[],
-            ),
-            clip=ClipField(
-                tokenizer=ModelInfo(
-                    key=model_key,
-                    submodel_type=SubModelType.Tokenizer,
-                ),
-                text_encoder=ModelInfo(
-                    key=model_key,
-                    submodel_type=SubModelType.TextEncoder,
-                ),
-                loras=[],
-                skipped_layers=0,
-            ),
-            clip2=ClipField(
-                tokenizer=ModelInfo(
-                    key=model_key,
-                    submodel_type=SubModelType.Tokenizer2,
-                ),
-                text_encoder=ModelInfo(
-                    key=model_key,
-                    submodel_type=SubModelType.TextEncoder2,
-                ),
-                loras=[],
-                skipped_layers=0,
-            ),
-            vae=VaeField(
-                vae=ModelInfo(
-                    key=model_key,
-                    submodel_type=SubModelType.Vae,
-                ),
-            ),
+            unet=UNetField(unet=unet, scheduler=scheduler, loras=[]),
+            clip=CLIPField(tokenizer=tokenizer, text_encoder=text_encoder, loras=[], skipped_layers=0),
+            clip2=CLIPField(tokenizer=tokenizer2, text_encoder=text_encoder2, loras=[], skipped_layers=0),
+            vae=VAEField(vae=vae),
         )
 
 
@@ -101,10 +72,8 @@ class SDXLModelLoaderInvocation(BaseInvocation):
 class SDXLRefinerModelLoaderInvocation(BaseInvocation):
     """Loads an sdxl refiner model, outputting its submodels."""
 
-    model: MainModelField = InputField(
-        description=FieldDescriptions.sdxl_refiner_model,
-        input=Input.Direct,
-        ui_type=UIType.SDXLRefinerModel,
+    model: ModelIdentifierField = InputField(
+        description=FieldDescriptions.sdxl_refiner_model, input=Input.Direct, ui_type=UIType.SDXLRefinerModel
     )
     # TODO: precision?
 
@@ -115,34 +84,14 @@ class SDXLRefinerModelLoaderInvocation(BaseInvocation):
         if not context.models.exists(model_key):
             raise Exception(f"Unknown model: {model_key}")
 
+        unet = self.model.model_copy(update={"submodel_type": SubModelType.UNet})
+        scheduler = self.model.model_copy(update={"submodel_type": SubModelType.Scheduler})
+        tokenizer2 = self.model.model_copy(update={"submodel_type": SubModelType.Tokenizer2})
+        text_encoder2 = self.model.model_copy(update={"submodel_type": SubModelType.TextEncoder2})
+        vae = self.model.model_copy(update={"submodel_type": SubModelType.VAE})
+
         return SDXLRefinerModelLoaderOutput(
-            unet=UNetField(
-                unet=ModelInfo(
-                    key=model_key,
-                    submodel_type=SubModelType.UNet,
-                ),
-                scheduler=ModelInfo(
-                    key=model_key,
-                    submodel_type=SubModelType.Scheduler,
-                ),
-                loras=[],
-            ),
-            clip2=ClipField(
-                tokenizer=ModelInfo(
-                    key=model_key,
-                    submodel_type=SubModelType.Tokenizer2,
-                ),
-                text_encoder=ModelInfo(
-                    key=model_key,
-                    submodel_type=SubModelType.TextEncoder2,
-                ),
-                loras=[],
-                skipped_layers=0,
-            ),
-            vae=VaeField(
-                vae=ModelInfo(
-                    key=model_key,
-                    submodel_type=SubModelType.Vae,
-                ),
-            ),
+            unet=UNetField(unet=unet, scheduler=scheduler, loras=[]),
+            clip2=CLIPField(tokenizer=tokenizer2, text_encoder=text_encoder2, loras=[], skipped_layers=0),
+            vae=VAEField(vae=vae),
         )

invokeai/app/invocations/t2i_adapter.py

@@ -9,18 +9,15 @@ from invokeai.app.invocations.baseinvocation import (
     invocation_output,
 )
 from invokeai.app.invocations.controlnet_image_processors import CONTROLNET_RESIZE_VALUES
-from invokeai.app.invocations.fields import FieldDescriptions, ImageField, Input, InputField, OutputField
+from invokeai.app.invocations.fields import FieldDescriptions, ImageField, Input, InputField, OutputField, UIType
+from invokeai.app.invocations.model import ModelIdentifierField
 from invokeai.app.invocations.util import validate_begin_end_step, validate_weights
 from invokeai.app.services.shared.invocation_context import InvocationContext
 
 
-class T2IAdapterModelField(BaseModel):
-    key: str = Field(description="Model record key for the T2I-Adapter model")
-
-
 class T2IAdapterField(BaseModel):
     image: ImageField = Field(description="The T2I-Adapter image prompt.")
-    t2i_adapter_model: T2IAdapterModelField = Field(description="The T2I-Adapter model to use.")
+    t2i_adapter_model: ModelIdentifierField = Field(description="The T2I-Adapter model to use.")
     weight: Union[float, list[float]] = Field(default=1, description="The weight given to the T2I-Adapter")
     begin_step_percent: float = Field(
         default=0, ge=0, le=1, description="When the T2I-Adapter is first applied (% of total steps)"
@@ -55,11 +52,12 @@ class T2IAdapterInvocation(BaseInvocation):
 
     # Inputs
     image: ImageField = InputField(description="The IP-Adapter image prompt.")
-    t2i_adapter_model: T2IAdapterModelField = InputField(
+    t2i_adapter_model: ModelIdentifierField = InputField(
         description="The T2I-Adapter model.",
         title="T2I-Adapter Model",
         input=Input.Direct,
         ui_order=-1,
+        ui_type=UIType.T2IAdapterModel,
     )
     weight: Union[float, list[float]] = InputField(
         default=1, ge=0, description="The weight given to the T2I-Adapter", title="Weight"

invokeai/app/services/config/config_base.py

@@ -17,7 +17,8 @@ from argparse import ArgumentParser
 from pathlib import Path
 from typing import Any, ClassVar, Dict, List, Literal, Optional, Union, get_args, get_origin, get_type_hints
 
-from omegaconf import DictConfig, ListConfig, OmegaConf
+from omegaconf import DictConfig, DictKeyType, ListConfig, OmegaConf
+from pydantic import BaseModel
 from pydantic_settings import BaseSettings, SettingsConfigDict
 
 from invokeai.app.services.config.config_common import PagingArgumentParser, int_or_float_or_str
@@ -62,6 +63,22 @@ class InvokeAISettings(BaseSettings):
             assert isinstance(category, str)
             if category not in field_dict[type]:
                 field_dict[type][category] = {}
+            if isinstance(value, BaseModel):
+                dump = value.model_dump(exclude_defaults=True, exclude_unset=True, exclude_none=True)
+                field_dict[type][category][name] = dump
+                continue
+            if isinstance(value, list):
+                if not value or len(value) == 0:
+                    continue
+                primitive = isinstance(value[0], get_args(DictKeyType))
+                if not primitive:
+                    val_list: List[Dict[str, Any]] = []
+                    for list_val in value:
+                        if isinstance(list_val, BaseModel):
+                            dump = list_val.model_dump(exclude_defaults=True, exclude_unset=True, exclude_none=True)
+                            val_list.append(dump)
+                    field_dict[type][category][name] = val_list
+                    continue
             # keep paths as strings to make it easier to read
             field_dict[type][category][name] = str(value) if isinstance(value, Path) else value
         conf = OmegaConf.create(field_dict)

invokeai/app/services/config/config_default.py

@@ -166,17 +166,21 @@ two configs are kept in separate sections of the config file:
      ...
 
 """
+
 from __future__ import annotations
 
 import os
+import re
 from pathlib import Path
 from typing import Any, ClassVar, Dict, List, Literal, Optional
 
 from omegaconf import DictConfig, OmegaConf
-from pydantic import Field
+from pydantic import BaseModel, Field, field_validator
 from pydantic.config import JsonDict
 from pydantic_settings import SettingsConfigDict
 
+from invokeai.backend.model_hash.model_hash import HASHING_ALGORITHMS
+
 from .config_base import InvokeAISettings
 
 INIT_FILE = Path("invokeai.yaml")
@@ -195,17 +199,87 @@ class Categories(object):
     Paths: JsonDict = {"category": "Paths"}
     Logging: JsonDict = {"category": "Logging"}
     Development: JsonDict = {"category": "Development"}
-    Other: JsonDict = {"category": "Other"}
+    CLIArgs: JsonDict = {"category": "CLIArgs"}
+    ModelInstall: JsonDict = {"category": "Model Install"}
     ModelCache: JsonDict = {"category": "Model Cache"}
     Device: JsonDict = {"category": "Device"}
     Generation: JsonDict = {"category": "Generation"}
     Queue: JsonDict = {"category": "Queue"}
     Nodes: JsonDict = {"category": "Nodes"}
     MemoryPerformance: JsonDict = {"category": "Memory/Performance"}
+    Deprecated: JsonDict = {"category": "Deprecated"}
+
+
+class URLRegexToken(BaseModel):
+    url_regex: str = Field(description="Regular expression to match against the URL")
+    token: str = Field(description="Token to use when the URL matches the regex")
+
+    @field_validator("url_regex")
+    @classmethod
+    def validate_url_regex(cls, v: str) -> str:
+        """Validate that the value is a valid regex."""
+        try:
+            re.compile(v)
+        except re.error as e:
+            raise ValueError(f"Invalid regex: {e}")
+        return v
 
 
 class InvokeAIAppConfig(InvokeAISettings):
-    """Configuration object for InvokeAI App."""
+    """Invoke App Configuration
+
+    Attributes:
+        host: **Web Server**: IP address to bind to. Use `0.0.0.0` to serve to your local network.
+        port: **Web Server**: Port to bind to.
+        allow_origins: **Web Server**: Allowed CORS origins.
+        allow_credentials: **Web Server**: Allow CORS credentials.
+        allow_methods: **Web Server**: Methods allowed for CORS.
+        allow_headers: **Web Server**: Headers allowed for CORS.
+        ssl_certfile: **Web Server**: SSL certificate file for HTTPS.
+        ssl_keyfile: **Web Server**: SSL key file for HTTPS.
+        esrgan: **Features**: Enables or disables the upscaling code.
+        internet_available: **Features**: If true, attempt to download models on the fly; otherwise only use local models.
+        log_tokenization: **Features**: Enable logging of parsed prompt tokens.
+        patchmatch: **Features**: Enable patchmatch inpaint code.
+        ignore_missing_core_models: **Features**: Ignore missing core models on startup. If `True`, the app will attempt to download missing models on startup.
+        root: **Paths**: The InvokeAI runtime root directory.
+        autoimport_dir: **Paths**: Path to a directory of models files to be imported on startup.
+        models_dir: **Paths**: Path to the models directory.
+        convert_cache_dir: **Paths**: Path to the converted models cache directory. When loading a non-diffusers model, it will be converted and store on disk at this location.
+        legacy_conf_dir: **Paths**: Path to directory of legacy checkpoint config files.
+        db_dir: **Paths**: Path to InvokeAI databases directory.
+        outdir: **Paths**: Path to directory for outputs.
+        custom_nodes_dir: **Paths**: Path to directory for custom nodes.
+        from_file: **Paths**: Take command input from the indicated file (command-line client only).
+        log_handlers: **Logging**: Log handler. Valid options are "console", "file=<path>", "syslog=path|address:host:port", "http=<url>".
+        log_format: **Logging**: Log format. Use "plain" for text-only, "color" for colorized output, "legacy" for 2.3-style logging and "syslog" for syslog-style.
+        log_level: **Logging**: Emit logging messages at this level or higher.
+        log_sql: **Logging**: Log SQL queries. `log_level` must be `debug` for this to do anything. Extremely verbose.
+        use_memory_db: **Development**: Use in-memory database. Useful for development.
+        dev_reload: **Development**: Automatically reload when Python sources are changed. Does not reload node definitions.
+        profile_graphs: **Development**: Enable graph profiling using `cProfile`.
+        profile_prefix: **Development**: An optional prefix for profile output files.
+        profiles_dir: **Development**: Path to profiles output directory.
+        version: **CLIArgs**: CLI arg - show InvokeAI version and exit.
+        hashing_algorithm: **Model Install**: Model hashing algorthim for model installs. 'blake3' is best for SSDs. 'blake3_single' is best for spinning disk HDDs. 'random' disables hashing, instead assigning a UUID to models. Useful when using a memory db to reduce model installation time, or if you don't care about storing stable hashes for models. Alternatively, any other hashlib algorithm is accepted, though these are not nearly as performant as blake3.
+        remote_api_tokens: **Model Install**: List of regular expression and token pairs used when downloading models from URLs. The download URL is tested against the regex, and if it matches, the token is provided in as a Bearer token.
+        ram: **Model Cache**: Maximum memory amount used by memory model cache for rapid switching (GB).
+        vram: **Model Cache**: Amount of VRAM reserved for model storage (GB)
+        convert_cache: **Model Cache**: Maximum size of on-disk converted models cache (GB)
+        lazy_offload: **Model Cache**: Keep models in VRAM until their space is needed.
+        log_memory_usage: **Model Cache**: If True, a memory snapshot will be captured before and after every model cache operation, and the result will be logged (at debug level). There is a time cost to capturing the memory snapshots, so it is recommended to only enable this feature if you are actively inspecting the model cache's behaviour.
+        device: **Device**: Preferred execution device. `auto` will choose the device depending on the hardware platform and the installed torch capabilities.
+        precision: **Device**: Floating point precision. `float16` will consume half the memory of `float32` but produce slightly lower-quality images. The `auto` setting will guess the proper precision based on your video card and operating system.
+        sequential_guidance: **Generation**: Whether to calculate guidance in serial instead of in parallel, lowering memory requirements.
+        attention_type: **Generation**: Attention type.
+        attention_slice_size: **Generation**: Slice size, valid when attention_type=="sliced".
+        force_tiled_decode: **Generation**: Whether to enable tiled VAE decode (reduces memory consumption with some performance penalty).
+        png_compress_level: **Generation**: The compress_level setting of PIL.Image.save(), used for PNG encoding. All settings are lossless. 0 = no compression, 1 = fastest with slightly larger filesize, 9 = slowest with smallest filesize. 1 is typically the best setting.
+        max_queue_size: **Queue**: Maximum number of items in the session queue.
+        allow_nodes: **Nodes**: List of nodes to allow. Omit to allow all.
+        deny_nodes: **Nodes**: List of nodes to deny. Omit to deny none.
+        node_cache_size: **Nodes**: How many cached nodes to keep in memory.
+    """
 
     singleton_config: ClassVar[Optional[InvokeAIAppConfig]] = None
     singleton_init: ClassVar[Optional[Dict[str, Any]]] = None
@@ -214,90 +288,98 @@ class InvokeAIAppConfig(InvokeAISettings):
     type: Literal["InvokeAI"] = "InvokeAI"
 
     # WEB
-    host                : str = Field(default="127.0.0.1", description="IP address to bind to", json_schema_extra=Categories.WebServer)
-    port                : int = Field(default=9090, description="Port to bind to", json_schema_extra=Categories.WebServer)
-    allow_origins       : List[str] = Field(default=[], description="Allowed CORS origins", json_schema_extra=Categories.WebServer)
-    allow_credentials   : bool = Field(default=True, description="Allow CORS credentials", json_schema_extra=Categories.WebServer)
-    allow_methods       : List[str] = Field(default=["*"], description="Methods allowed for CORS", json_schema_extra=Categories.WebServer)
-    allow_headers       : List[str] = Field(default=["*"], description="Headers allowed for CORS", json_schema_extra=Categories.WebServer)
+    host                : str = Field(default="127.0.0.1", description="IP address to bind to. Use `0.0.0.0` to serve to your local network.", json_schema_extra=Categories.WebServer)
+    port                : int = Field(default=9090, description="Port to bind to.", json_schema_extra=Categories.WebServer)
+    allow_origins       : List[str] = Field(default=[], description="Allowed CORS origins.", json_schema_extra=Categories.WebServer)
+    allow_credentials   : bool = Field(default=True, description="Allow CORS credentials.", json_schema_extra=Categories.WebServer)
+    allow_methods       : List[str] = Field(default=["*"], description="Methods allowed for CORS.", json_schema_extra=Categories.WebServer)
+    allow_headers       : List[str] = Field(default=["*"], description="Headers allowed for CORS.", json_schema_extra=Categories.WebServer)
     # SSL options correspond to https://www.uvicorn.org/settings/#https
-    ssl_certfile        : Optional[Path] = Field(default=None, description="SSL certificate file (for HTTPS)", json_schema_extra=Categories.WebServer)
-    ssl_keyfile         : Optional[Path] = Field(default=None, description="SSL key file", json_schema_extra=Categories.WebServer)
+    ssl_certfile        : Optional[Path] = Field(default=None, description="SSL certificate file for HTTPS.", json_schema_extra=Categories.WebServer)
+    ssl_keyfile         : Optional[Path] = Field(default=None, description="SSL key file for HTTPS.", json_schema_extra=Categories.WebServer)
 
     # FEATURES
-    esrgan              : bool = Field(default=True, description="Enable/disable upscaling code", json_schema_extra=Categories.Features)
-    internet_available  : bool = Field(default=True, description="If true, attempt to download models on the fly; otherwise only use local models", json_schema_extra=Categories.Features)
+    esrgan              : bool = Field(default=True, description="Enables or disables the upscaling code.", json_schema_extra=Categories.Features)
+    # TODO(psyche): This is not used anywhere.
+    internet_available  : bool = Field(default=True, description="If true, attempt to download models on the fly; otherwise only use local models.", json_schema_extra=Categories.Features)
     log_tokenization    : bool = Field(default=False, description="Enable logging of parsed prompt tokens.", json_schema_extra=Categories.Features)
-    patchmatch          : bool = Field(default=True, description="Enable/disable patchmatch inpaint code", json_schema_extra=Categories.Features)
-    ignore_missing_core_models : bool = Field(default=False, description='Ignore missing models in models/core/convert', json_schema_extra=Categories.Features)
+    patchmatch          : bool = Field(default=True, description="Enable patchmatch inpaint code.", json_schema_extra=Categories.Features)
+    ignore_missing_core_models : bool = Field(default=False, description='Ignore missing core models on startup. If `True`, the app will attempt to download missing models on startup.', json_schema_extra=Categories.Features)
 
     # PATHS
-    root                : Optional[Path] = Field(default=None, description='InvokeAI runtime root directory', json_schema_extra=Categories.Paths)
+    root                : Optional[Path] = Field(default=None, description='The InvokeAI runtime root directory.', json_schema_extra=Categories.Paths)
     autoimport_dir      : Path = Field(default=Path('autoimport'), description='Path to a directory of models files to be imported on startup.', json_schema_extra=Categories.Paths)
-    models_dir          : Path = Field(default=Path('models'), description='Path to the models directory', json_schema_extra=Categories.Paths)
-    convert_cache_dir   : Path = Field(default=Path('models/.cache'), description='Path to the converted models cache directory', json_schema_extra=Categories.Paths)
-    legacy_conf_dir     : Path = Field(default=Path('configs/stable-diffusion'), description='Path to directory of legacy checkpoint config files', json_schema_extra=Categories.Paths)
-    db_dir              : Path = Field(default=Path('databases'), description='Path to InvokeAI databases directory', json_schema_extra=Categories.Paths)
-    outdir              : Path = Field(default=Path('outputs'), description='Default folder for output images', json_schema_extra=Categories.Paths)
-    use_memory_db       : bool = Field(default=False, description='Use in-memory database for storing image metadata', json_schema_extra=Categories.Paths)
-    custom_nodes_dir    : Path = Field(default=Path('nodes'), description='Path to directory for custom nodes', json_schema_extra=Categories.Paths)
-    from_file           : Optional[Path] = Field(default=None, description='Take command input from the indicated file (command-line client only)', json_schema_extra=Categories.Paths)
+    models_dir          : Path = Field(default=Path('models'), description='Path to the models directory.', json_schema_extra=Categories.Paths)
+    convert_cache_dir   : Path = Field(default=Path('models/.cache'), description='Path to the converted models cache directory. When loading a non-diffusers model, it will be converted and store on disk at this location.', json_schema_extra=Categories.Paths)
+    legacy_conf_dir     : Path = Field(default=Path('configs/stable-diffusion'), description='Path to directory of legacy checkpoint config files.', json_schema_extra=Categories.Paths)
+    db_dir              : Path = Field(default=Path('databases'), description='Path to InvokeAI databases directory.', json_schema_extra=Categories.Paths)
+    outdir              : Path = Field(default=Path('outputs'), description='Path to directory for outputs.', json_schema_extra=Categories.Paths)
+    custom_nodes_dir    : Path = Field(default=Path('nodes'), description='Path to directory for custom nodes.', json_schema_extra=Categories.Paths)
+    # TODO(psyche): This is not used anywhere.
+    from_file           : Optional[Path] = Field(default=None, description='Take command input from the indicated file (command-line client only).', json_schema_extra=Categories.Paths)
 
     # LOGGING
-    log_handlers        : List[str] = Field(default=["console"], description='Log handler. Valid options are "console", "file=<path>", "syslog=path|address:host:port", "http=<url>"', json_schema_extra=Categories.Logging)
+    log_handlers        : List[str] = Field(default=["console"], description='Log handler. Valid options are "console", "file=<path>", "syslog=path|address:host:port", "http=<url>".', json_schema_extra=Categories.Logging)
     # note - would be better to read the log_format values from logging.py, but this creates circular dependencies issues
-    log_format          : Literal['plain', 'color', 'syslog', 'legacy'] = Field(default="color", description='Log format. Use "plain" for text-only, "color" for colorized output, "legacy" for 2.3-style logging and "syslog" for syslog-style', json_schema_extra=Categories.Logging)
-    log_level           : Literal["debug", "info", "warning", "error", "critical"] = Field(default="info", description="Emit logging messages at this level or  higher", json_schema_extra=Categories.Logging)
-    log_sql             : bool = Field(default=False, description="Log SQL queries", json_schema_extra=Categories.Logging)
+    log_format          : Literal['plain', 'color', 'syslog', 'legacy'] = Field(default="color", description='Log format. Use "plain" for text-only, "color" for colorized output, "legacy" for 2.3-style logging and "syslog" for syslog-style.', json_schema_extra=Categories.Logging)
+    log_level           : Literal["debug", "info", "warning", "error", "critical"] = Field(default="info", description="Emit logging messages at this level or higher.", json_schema_extra=Categories.Logging)
+    log_sql             : bool = Field(default=False, description="Log SQL queries. `log_level` must be `debug` for this to do anything. Extremely verbose.", json_schema_extra=Categories.Logging)
 
     # Development
-    dev_reload          : bool = Field(default=False, description="Automatically reload when Python sources are changed.", json_schema_extra=Categories.Development)
-    profile_graphs      : bool = Field(default=False, description="Enable graph profiling", json_schema_extra=Categories.Development)
+    use_memory_db       : bool = Field(default=False, description='Use in-memory database. Useful for development.', json_schema_extra=Categories.Development)
+    dev_reload          : bool = Field(default=False, description="Automatically reload when Python sources are changed. Does not reload node definitions.", json_schema_extra=Categories.Development)
+    profile_graphs      : bool = Field(default=False, description="Enable graph profiling using `cProfile`.", json_schema_extra=Categories.Development)
     profile_prefix      : Optional[str] = Field(default=None, description="An optional prefix for profile output files.", json_schema_extra=Categories.Development)
-    profiles_dir        : Path = Field(default=Path('profiles'), description="Directory for graph profiles", json_schema_extra=Categories.Development)
+    profiles_dir        : Path = Field(default=Path('profiles'), description="Path to profiles output directory.", json_schema_extra=Categories.Development)
 
-    version             : bool = Field(default=False, description="Show InvokeAI version and exit", json_schema_extra=Categories.Other)
+    version             : bool = Field(default=False, description="CLI arg - show InvokeAI version and exit.", json_schema_extra=Categories.CLIArgs)
 
     # CACHE
-    ram                 : float = Field(default=DEFAULT_RAM_CACHE, gt=0, description="Maximum memory amount used by model cache for rapid switching (floating point number, GB)", json_schema_extra=Categories.ModelCache, )
-    vram                : float = Field(default=DEFAULT_VRAM_CACHE, ge=0, description="Amount of VRAM reserved for model storage (floating point number, GB)", json_schema_extra=Categories.ModelCache, )
+    ram                 : float = Field(default=DEFAULT_RAM_CACHE, gt=0, description="Maximum memory amount used by memory model cache for rapid switching (GB).", json_schema_extra=Categories.ModelCache, )
+    vram                : float = Field(default=DEFAULT_VRAM_CACHE, ge=0, description="Amount of VRAM reserved for model storage (GB)", json_schema_extra=Categories.ModelCache, )
     convert_cache       : float = Field(default=DEFAULT_CONVERT_CACHE, ge=0, description="Maximum size of on-disk converted models cache (GB)", json_schema_extra=Categories.ModelCache)
 
-    lazy_offload        : bool = Field(default=True, description="Keep models in VRAM until their space is needed", json_schema_extra=Categories.ModelCache, )
+    lazy_offload        : bool = Field(default=True, description="Keep models in VRAM until their space is needed.", json_schema_extra=Categories.ModelCache, )
     log_memory_usage    : bool = Field(default=False, description="If True, a memory snapshot will be captured before and after every model cache operation, and the result will be logged (at debug level). There is a time cost to capturing the memory snapshots, so it is recommended to only enable this feature if you are actively inspecting the model cache's behaviour.", json_schema_extra=Categories.ModelCache)
 
     # DEVICE
-    device              : Literal["auto", "cpu", "cuda", "cuda:1", "mps"] = Field(default="auto", description="Generation device", json_schema_extra=Categories.Device)
-    precision           : Literal["auto", "float16", "bfloat16", "float32", "autocast"] = Field(default="auto", description="Floating point precision", json_schema_extra=Categories.Device)
+    device              : Literal["auto", "cpu", "cuda", "cuda:1", "mps"] = Field(default="auto", description="Preferred execution device. `auto` will choose the device depending on the hardware platform and the installed torch capabilities.", json_schema_extra=Categories.Device)
+    precision           : Literal["auto", "float16", "bfloat16", "float32", "autocast"] = Field(default="auto", description="Floating point precision. `float16` will consume half the memory of `float32` but produce slightly lower-quality images. The `auto` setting will guess the proper precision based on your video card and operating system.", json_schema_extra=Categories.Device)
 
     # GENERATION
-    sequential_guidance : bool = Field(default=False, description="Whether to calculate guidance in serial instead of in parallel, lowering memory requirements", json_schema_extra=Categories.Generation)
-    attention_type      : Literal["auto", "normal", "xformers", "sliced", "torch-sdp"] = Field(default="auto", description="Attention type", json_schema_extra=Categories.Generation)
-    attention_slice_size: Literal["auto", "balanced", "max", 1, 2, 3, 4, 5, 6, 7, 8] = Field(default="auto", description='Slice size, valid when attention_type=="sliced"', json_schema_extra=Categories.Generation)
-    force_tiled_decode  : bool = Field(default=False, description="Whether to enable tiled VAE decode (reduces memory consumption with some performance penalty)", json_schema_extra=Categories.Generation)
-    png_compress_level  : int = Field(default=1, description="The compress_level setting of PIL.Image.save(), used for PNG encoding. All settings are lossless. 0 = fastest, largest filesize, 9 = slowest, smallest filesize", json_schema_extra=Categories.Generation)
+    sequential_guidance : bool = Field(default=False, description="Whether to calculate guidance in serial instead of in parallel, lowering memory requirements.", json_schema_extra=Categories.Generation)
+    attention_type      : Literal["auto", "normal", "xformers", "sliced", "torch-sdp"] = Field(default="auto", description="Attention type.", json_schema_extra=Categories.Generation)
+    attention_slice_size: Literal["auto", "balanced", "max", 1, 2, 3, 4, 5, 6, 7, 8] = Field(default="auto", description='Slice size, valid when attention_type=="sliced".', json_schema_extra=Categories.Generation)
+    force_tiled_decode  : bool = Field(default=False, description="Whether to enable tiled VAE decode (reduces memory consumption with some performance penalty).", json_schema_extra=Categories.Generation)
+    png_compress_level  : int = Field(default=1, description="The compress_level setting of PIL.Image.save(), used for PNG encoding. All settings are lossless. 0 = no compression, 1 = fastest with slightly larger filesize, 9 = slowest with smallest filesize. 1 is typically the best setting.", json_schema_extra=Categories.Generation)
 
     # QUEUE
-    max_queue_size      : int = Field(default=10000, gt=0, description="Maximum number of items in the session queue", json_schema_extra=Categories.Queue)
+    max_queue_size      : int = Field(default=10000, gt=0, description="Maximum number of items in the session queue.", json_schema_extra=Categories.Queue)
 
     # NODES
     allow_nodes         : Optional[List[str]] = Field(default=None, description="List of nodes to allow. Omit to allow all.", json_schema_extra=Categories.Nodes)
     deny_nodes          : Optional[List[str]] = Field(default=None, description="List of nodes to deny. Omit to deny none.", json_schema_extra=Categories.Nodes)
-    node_cache_size     : int = Field(default=512, description="How many cached nodes to keep in memory", json_schema_extra=Categories.Nodes)
+    node_cache_size     : int = Field(default=512, description="How many cached nodes to keep in memory.", json_schema_extra=Categories.Nodes)
 
-    # MODEL IMPORT
-    civitai_api_key       : Optional[str] = Field(default=os.environ.get("CIVITAI_API_KEY"), description="API key for CivitAI", json_schema_extra=Categories.Other)
+    # MODEL INSTALL
+    hashing_algorithm   : HASHING_ALGORITHMS = Field(default="blake3", description="Model hashing algorthim for model installs. 'blake3' is best for SSDs. 'blake3_single' is best for spinning disk HDDs. 'random' disables hashing, instead assigning a UUID to models. Useful when using a memory db to reduce model installation time, or if you don't care about storing stable hashes for models. Alternatively, any other hashlib algorithm is accepted, though these are not nearly as performant as blake3.", json_schema_extra=Categories.ModelInstall)
+    remote_api_tokens   : Optional[list[URLRegexToken]] = Field(
+        default=None,
+        description="List of regular expression and token pairs used when downloading models from URLs. The download URL is tested against the regex, and if it matches, the token is provided in as a Bearer token.",
+        json_schema_extra=Categories.ModelInstall
+    )
 
+    # TODO(psyche): Can we just remove these then?
     # DEPRECATED FIELDS - STILL HERE IN ORDER TO OBTAN VALUES FROM PRE-3.1 CONFIG FILES
-    always_use_cpu      : bool = Field(default=False, description="If true, use the CPU for rendering even if a GPU is available.", json_schema_extra=Categories.MemoryPerformance)
-    max_cache_size      : Optional[float] = Field(default=None, gt=0, description="Maximum memory amount used by model cache for rapid switching", json_schema_extra=Categories.MemoryPerformance)
-    max_vram_cache_size : Optional[float] = Field(default=None, ge=0, description="Amount of VRAM reserved for model storage", json_schema_extra=Categories.MemoryPerformance)
-    xformers_enabled    : bool = Field(default=True, description="Enable/disable memory-efficient attention", json_schema_extra=Categories.MemoryPerformance)
-    tiled_decode        : bool = Field(default=False, description="Whether to enable tiled VAE decode (reduces memory consumption with some performance penalty)", json_schema_extra=Categories.MemoryPerformance)
-    lora_dir            : Optional[Path] = Field(default=None, description='Path to a directory of LoRA/LyCORIS models to be imported on startup.', json_schema_extra=Categories.Paths)
-    embedding_dir       : Optional[Path] = Field(default=None, description='Path to a directory of Textual Inversion embeddings to be imported on startup.', json_schema_extra=Categories.Paths)
-    controlnet_dir      : Optional[Path] = Field(default=None, description='Path to a directory of ControlNet embeddings to be imported on startup.', json_schema_extra=Categories.Paths)
-    conf_path           : Path = Field(default=Path('configs/models.yaml'), description='Path to models definition file', json_schema_extra=Categories.Paths)
+    always_use_cpu      : bool = Field(default=False, description="If true, use the CPU for rendering even if a GPU is available.", json_schema_extra=Categories.Deprecated)
+    max_cache_size      : Optional[float] = Field(default=None, gt=0, description="Maximum memory amount used by model cache for rapid switching", json_schema_extra=Categories.Deprecated)
+    max_vram_cache_size : Optional[float] = Field(default=None, ge=0, description="Amount of VRAM reserved for model storage", json_schema_extra=Categories.Deprecated)
+    xformers_enabled    : bool = Field(default=True, description="Enable/disable memory-efficient attention", json_schema_extra=Categories.Deprecated)
+    tiled_decode        : bool = Field(default=False, description="Whether to enable tiled VAE decode (reduces memory consumption with some performance penalty)", json_schema_extra=Categories.Deprecated)
+    lora_dir            : Optional[Path] = Field(default=None, description='Path to a directory of LoRA/LyCORIS models to be imported on startup.', json_schema_extra=Categories.Deprecated)
+    embedding_dir       : Optional[Path] = Field(default=None, description='Path to a directory of Textual Inversion embeddings to be imported on startup.', json_schema_extra=Categories.Deprecated)
+    controlnet_dir      : Optional[Path] = Field(default=None, description='Path to a directory of ControlNet embeddings to be imported on startup.', json_schema_extra=Categories.Deprecated)
+    conf_path           : Path = Field(default=Path('configs/models.yaml'), description='Path to models definition file', json_schema_extra=Categories.Deprecated)
 
     # this is not referred to in the source code and can be removed entirely
     #free_gpu_mem        : Optional[bool] = Field(default=None, description="If true, purge model from GPU after each generation.", json_schema_extra=Categories.MemoryPerformance)
@@ -475,6 +557,53 @@ class InvokeAIAppConfig(InvokeAISettings):
         """Choose the runtime root directory when not specified on command line or init file."""
         return _find_root()
 
+    @staticmethod
+    def generate_docstrings() -> str:
+        """Helper function for mkdocs. Generates a docstring for the InvokeAIAppConfig class.
+
+        You shouldn't run this manually. Instead, run `scripts/update-config-docstring.py` to update the docstring.
+        A makefile target is also available: `make update-config-docstring`.
+
+        See that script for more information about why this is necessary.
+        """
+        docstring = '    """Invoke App Configuration\n\n'
+        docstring += "    Attributes:"
+
+        field_descriptions: dict[str, list[str]] = {}
+
+        for k, v in InvokeAIAppConfig.model_fields.items():
+            if not isinstance(v.json_schema_extra, dict):
+                # Should never happen
+                continue
+
+            category = v.json_schema_extra.get("category", None)
+            if not isinstance(category, str) or category == "Deprecated":
+                continue
+            if not field_descriptions.get(category):
+                field_descriptions[category] = []
+            field_descriptions[category].append(f"        {k}: **{category}**: {v.description}")
+
+        for c in [
+            "Web Server",
+            "Features",
+            "Paths",
+            "Logging",
+            "Development",
+            "CLIArgs",
+            "Model Install",
+            "Model Cache",
+            "Device",
+            "Generation",
+            "Queue",
+            "Nodes",
+        ]:
+            docstring += "\n"
+            docstring += "\n".join(field_descriptions[c])
+
+        docstring += '\n    """'
+
+        return docstring
+
 
 def get_invokeai_config(**kwargs: Any) -> InvokeAIAppConfig:
     """Legacy function which returns InvokeAIAppConfig.get_config()."""

invokeai/app/services/download/__init__.py

@@ -1,4 +1,5 @@
 """Init file for download queue."""
+
 from .download_base import DownloadJob, DownloadJobStatus, DownloadQueueServiceBase, UnknownJobIDException
 from .download_default import DownloadQueueService, TqdmProgress
 

invokeai/app/services/download/download_default.py

@@ -224,20 +224,13 @@ class DownloadQueueService(DownloadQueueServiceBase):
                 job.job_started = get_iso_timestamp()
                 self._do_download(job)
                 self._signal_job_complete(job)
-
             except (OSError, HTTPError) as excp:
                 job.error_type = excp.__class__.__name__ + f"({str(excp)})"
                 job.error = traceback.format_exc()
-                try:
                 self._signal_job_error(job, excp)
-                except:
-                    pass
             except DownloadJobCancelledException:
-                try:
                 self._signal_job_cancelled(job)
                 self._cleanup_cancelled_job(job)
-                except:
-                    pass
 
             finally:
                 job.job_ended = get_iso_timestamp()

invokeai/app/services/events/events_base.py

@@ -12,6 +12,7 @@ from invokeai.app.services.session_queue.session_queue_common import (
 )
 from invokeai.app.util.misc import get_timestamp
 from invokeai.backend.model_manager import AnyModelConfig
+from invokeai.backend.model_manager.config import SubModelType
 
 
 class EventServiceBase:
@@ -80,7 +81,7 @@ class EventServiceBase:
                 "graph_execution_state_id": graph_execution_state_id,
                 "node_id": node_id,
                 "source_node_id": source_node_id,
-                "progress_image": progress_image.model_dump() if progress_image is not None else None,
+                "progress_image": progress_image.model_dump(mode="json") if progress_image is not None else None,
                 "step": step,
                 "order": order,
                 "total_steps": total_steps,
@@ -180,6 +181,7 @@ class EventServiceBase:
         queue_batch_id: str,
         graph_execution_state_id: str,
         model_config: AnyModelConfig,
+        submodel_type: Optional[SubModelType] = None,
     ) -> None:
         """Emitted when a model is requested"""
         self.__emit_queue_event(
@@ -189,7 +191,8 @@ class EventServiceBase:
                 "queue_item_id": queue_item_id,
                 "queue_batch_id": queue_batch_id,
                 "graph_execution_state_id": graph_execution_state_id,
-                "model_config": model_config.model_dump(),
+                "model_config": model_config.model_dump(mode="json"),
+                "submodel_type": submodel_type,
             },
         )
 
@@ -200,6 +203,7 @@ class EventServiceBase:
         queue_batch_id: str,
         graph_execution_state_id: str,
         model_config: AnyModelConfig,
+        submodel_type: Optional[SubModelType] = None,
     ) -> None:
         """Emitted when a model is correctly loaded (returns model info)"""
         self.__emit_queue_event(
@@ -209,7 +213,8 @@ class EventServiceBase:
                 "queue_item_id": queue_item_id,
                 "queue_batch_id": queue_batch_id,
                 "graph_execution_state_id": graph_execution_state_id,
-                "model_config": model_config.model_dump(),
+                "model_config": model_config.model_dump(mode="json"),
+                "submodel_type": submodel_type,
             },
         )
 
@@ -254,8 +259,8 @@ class EventServiceBase:
                     "started_at": str(session_queue_item.started_at) if session_queue_item.started_at else None,
                     "completed_at": str(session_queue_item.completed_at) if session_queue_item.completed_at else None,
                 },
-                "batch_status": batch_status.model_dump(),
-                "queue_status": queue_status.model_dump(),
+                "batch_status": batch_status.model_dump(mode="json"),
+                "queue_status": queue_status.model_dump(mode="json"),
             },
         )
 
@@ -405,7 +410,7 @@ class EventServiceBase:
             payload={"source": source, "total_bytes": total_bytes, "key": key, "id": id},
         )
 
-    def emit_model_install_cancelled(self, source: str) -> None:
+    def emit_model_install_cancelled(self, source: str, id: int) -> None:
         """
         Emit when an install job is cancelled.
 
@@ -413,7 +418,7 @@ class EventServiceBase:
         """
         self.__emit_model_event(
             event_name="model_install_cancelled",
-            payload={"source": source},
+            payload={"source": source, "id": id},
         )
 
     def emit_model_install_error(self, source: str, error_type: str, error: str, id: int) -> None:

invokeai/app/services/invocation_cache/invocation_cache_base.py

@@ -41,8 +41,9 @@ class InvocationCacheBase(ABC):
         """Clears the cache"""
         pass
 
+    @staticmethod
     @abstractmethod
-    def create_key(self, invocation: BaseInvocation) -> int:
+    def create_key(invocation: BaseInvocation) -> int:
         """Gets the key for the invocation's cache item"""
         pass
 

invokeai/app/services/invocation_cache/invocation_cache_memory.py

@@ -61,9 +61,7 @@ class MemoryInvocationCache(InvocationCacheBase):
             self._delete_oldest_access(number_to_delete)
             self._cache[key] = CachedItem(
                 invocation_output,
-                invocation_output.model_dump_json(
-                    warnings=False, exclude_defaults=True, exclude_unset=True, include={"type"}
-                ),
+                invocation_output.model_dump_json(warnings=False, exclude_defaults=True, exclude_unset=True),
             )
 
     def _delete_oldest_access(self, number_to_delete: int) -> None:
@@ -81,7 +79,7 @@ class MemoryInvocationCache(InvocationCacheBase):
         with self._lock:
             return self._delete(key)
 
-    def clear(self, *args, **kwargs) -> None:
+    def clear(self) -> None:
         with self._lock:
             if self._max_cache_size == 0:
                 return

invokeai/app/services/invocation_services.py

@@ -25,6 +25,7 @@ if TYPE_CHECKING:
     from .images.images_base import ImageServiceABC
     from .invocation_cache.invocation_cache_base import InvocationCacheBase
     from .invocation_stats.invocation_stats_base import InvocationStatsServiceBase
+    from .model_images.model_images_base import ModelImageFileStorageBase
     from .model_manager.model_manager_base import ModelManagerServiceBase
     from .names.names_base import NameServiceBase
     from .session_processor.session_processor_base import SessionProcessorBase
@@ -49,6 +50,7 @@ class InvocationServices:
         image_files: "ImageFileStorageBase",
         image_records: "ImageRecordStorageBase",
         logger: "Logger",
+        model_images: "ModelImageFileStorageBase",
         model_manager: "ModelManagerServiceBase",
         download_queue: "DownloadQueueServiceBase",
         performance_statistics: "InvocationStatsServiceBase",
@@ -72,6 +74,7 @@ class InvocationServices:
         self.image_files = image_files
         self.image_records = image_records
         self.logger = logger
+        self.model_images = model_images
         self.model_manager = model_manager
         self.download_queue = download_queue
         self.performance_statistics = performance_statistics

invokeai/app/services/model_images/model_images_base.py

@@ -0,0 +1,33 @@
+from abc import ABC, abstractmethod
+from pathlib import Path
+
+from PIL.Image import Image as PILImageType
+
+
+class ModelImageFileStorageBase(ABC):
+    """Low-level service responsible for storing and retrieving image files."""
+
+    @abstractmethod
+    def get(self, model_key: str) -> PILImageType:
+        """Retrieves a model image as PIL Image."""
+        pass
+
+    @abstractmethod
+    def get_path(self, model_key: str) -> Path:
+        """Gets the internal path to a model image."""
+        pass
+
+    @abstractmethod
+    def get_url(self, model_key: str) -> str | None:
+        """Gets the URL to fetch a model image."""
+        pass
+
+    @abstractmethod
+    def save(self, image: PILImageType, model_key: str) -> None:
+        """Saves a model image."""
+        pass
+
+    @abstractmethod
+    def delete(self, model_key: str) -> None:
+        """Deletes a model image."""
+        pass

invokeai/app/services/model_images/model_images_common.py

@@ -0,0 +1,20 @@
+# TODO: Should these excpetions subclass existing python exceptions?
+class ModelImageFileNotFoundException(Exception):
+    """Raised when an image file is not found in storage."""
+
+    def __init__(self, message="Model image file not found"):
+        super().__init__(message)
+
+
+class ModelImageFileSaveException(Exception):
+    """Raised when an image cannot be saved."""
+
+    def __init__(self, message="Model image file not saved"):
+        super().__init__(message)
+
+
+class ModelImageFileDeleteException(Exception):
+    """Raised when an image cannot be deleted."""
+
+    def __init__(self, message="Model image file not deleted"):
+        super().__init__(message)

invokeai/app/services/model_images/model_images_default.py

@@ -0,0 +1,85 @@
+from pathlib import Path
+
+from PIL import Image
+from PIL.Image import Image as PILImageType
+from send2trash import send2trash
+
+from invokeai.app.services.invoker import Invoker
+from invokeai.app.util.misc import uuid_string
+from invokeai.app.util.thumbnails import make_thumbnail
+
+from .model_images_base import ModelImageFileStorageBase
+from .model_images_common import (
+    ModelImageFileDeleteException,
+    ModelImageFileNotFoundException,
+    ModelImageFileSaveException,
+)
+
+
+class ModelImageFileStorageDisk(ModelImageFileStorageBase):
+    """Stores images on disk"""
+
+    def __init__(self, model_images_folder: Path):
+        self._model_images_folder = model_images_folder
+        self._validate_storage_folders()
+
+    def start(self, invoker: Invoker) -> None:
+        self._invoker = invoker
+
+    def get(self, model_key: str) -> PILImageType:
+        try:
+            path = self.get_path(model_key)
+
+            if not self._validate_path(path):
+                raise ModelImageFileNotFoundException
+
+            return Image.open(path)
+        except FileNotFoundError as e:
+            raise ModelImageFileNotFoundException from e
+
+    def save(self, image: PILImageType, model_key: str) -> None:
+        try:
+            self._validate_storage_folders()
+            image_path = self._model_images_folder / (model_key + ".webp")
+            thumbnail = make_thumbnail(image, 256)
+            thumbnail.save(image_path, format="webp")
+
+        except Exception as e:
+            raise ModelImageFileSaveException from e
+
+    def get_path(self, model_key: str) -> Path:
+        path = self._model_images_folder / (model_key + ".webp")
+
+        return path
+
+    def get_url(self, model_key: str) -> str | None:
+        path = self.get_path(model_key)
+        if not self._validate_path(path):
+            return
+
+        url = self._invoker.services.urls.get_model_image_url(model_key)
+
+        # The image URL never changes, so we must add random query string to it to prevent caching
+        url += f"?{uuid_string()}"
+
+        return url
+
+    def delete(self, model_key: str) -> None:
+        try:
+            path = self.get_path(model_key)
+
+            if not self._validate_path(path):
+                raise ModelImageFileNotFoundException
+
+            send2trash(path)
+
+        except Exception as e:
+            raise ModelImageFileDeleteException from e
+
+    def _validate_path(self, path: Path) -> bool:
+        """Validates the path given for an image."""
+        return path.exists()
+
+    def _validate_storage_folders(self) -> None:
+        """Checks if the required folders exist and create them if they don't"""
+        self._model_images_folder.mkdir(parents=True, exist_ok=True)

invokeai/app/services/model_install/__init__.py

@@ -1,7 +1,6 @@
 """Initialization file for model install service package."""
 
 from .model_install_base import (
-    CivitaiModelSource,
     HFModelSource,
     InstallStatus,
     LocalModelSource,
@@ -23,5 +22,4 @@ __all__ = [
     "LocalModelSource",
     "HFModelSource",
     "URLModelSource",
-    "CivitaiModelSource",
 ]

invokeai/app/services/model_install/model_install_base.py

@@ -18,16 +18,16 @@ from invokeai.app.services.events.events_base import EventServiceBase
 from invokeai.app.services.invoker import Invoker
 from invokeai.app.services.model_records import ModelRecordServiceBase
 from invokeai.backend.model_manager import AnyModelConfig, ModelRepoVariant
+from invokeai.backend.model_manager.config import ModelSourceType
 from invokeai.backend.model_manager.metadata import AnyModelRepoMetadata
 
-from ..model_metadata import ModelMetadataStoreBase
-
 
 class InstallStatus(str, Enum):
     """State of an install job running in the background."""
 
     WAITING = "waiting"  # waiting to be dequeued
     DOWNLOADING = "downloading"  # downloading of model files in process
+    DOWNLOADS_DONE = "downloads_done"  # downloading done, waiting to run
     RUNNING = "running"  # being processed
     COMPLETED = "completed"  # finished running
     ERROR = "error"  # terminated with an error message
@@ -91,21 +91,6 @@ class LocalModelSource(StringLikeSource):
         return Path(self.path).as_posix()
 
 
-class CivitaiModelSource(StringLikeSource):
-    """A Civitai version id, with optional variant and access token."""
-
-    version_id: int
-    variant: Optional[ModelRepoVariant] = None
-    access_token: Optional[str] = None
-    type: Literal["civitai"] = "civitai"
-
-    def __str__(self) -> str:
-        """Return string version of repoid when string rep needed."""
-        base: str = str(self.version_id)
-        base += f" ({self.variant})" if self.variant else ""
-        return base
-
-
 class HFModelSource(StringLikeSource):
     """
     A HuggingFace repo_id with optional variant, sub-folder and access token.
@@ -146,9 +131,13 @@ class URLModelSource(StringLikeSource):
         return str(self.url)
 
 
-ModelSource = Annotated[
-    Union[LocalModelSource, HFModelSource, CivitaiModelSource, URLModelSource], Field(discriminator="type")
-]
+ModelSource = Annotated[Union[LocalModelSource, HFModelSource, URLModelSource], Field(discriminator="type")]
+
+MODEL_SOURCE_TO_TYPE_MAP = {
+    URLModelSource: ModelSourceType.Url,
+    HFModelSource: ModelSourceType.HFRepoID,
+    LocalModelSource: ModelSourceType.Path,
+}
 
 
 class ModelInstallJob(BaseModel):
@@ -229,6 +218,11 @@ class ModelInstallJob(BaseModel):
         """Return true if job is downloading."""
         return self.status == InstallStatus.DOWNLOADING
 
+    @property
+    def downloads_done(self) -> bool:
+        """Return true if job's downloads ae done."""
+        return self.status == InstallStatus.DOWNLOADS_DONE
+
     @property
     def running(self) -> bool:
         """Return true if job is running."""
@@ -254,7 +248,6 @@ class ModelInstallServiceBase(ABC):
         app_config: InvokeAIAppConfig,
         record_store: ModelRecordServiceBase,
         download_queue: DownloadQueueServiceBase,
-        metadata_store: ModelMetadataStoreBase,
         event_bus: Optional["EventServiceBase"] = None,
     ):
         """
@@ -341,6 +334,7 @@ class ModelInstallServiceBase(ABC):
         source: str,
         config: Optional[Dict[str, Any]] = None,
         access_token: Optional[str] = None,
+        inplace: Optional[bool] = False,
     ) -> ModelInstallJob:
         r"""Install the indicated model using heuristics to interpret user intentions.
 
@@ -386,7 +380,7 @@ class ModelInstallServiceBase(ABC):
          will override corresponding autoassigned probe fields in the
          model's config record. Use it to override
          `name`, `description`, `base_type`, `model_type`, `format`,
-         `prediction_type`, `image_size`, and/or `ztsnr_training`.
+         `prediction_type`, and/or `image_size`.
 
         This will download the model located at `source`,
         probe it, and install it into the models directory.

invokeai/app/services/model_install/model_install_default.py

@@ -7,12 +7,12 @@ import time
 from hashlib import sha256
 from pathlib import Path
 from queue import Empty, Queue
-from random import randbytes
 from shutil import copyfile, copytree, move, rmtree
 from tempfile import mkdtemp
 from typing import Any, Dict, List, Optional, Set, Union
 
 from huggingface_hub import HfFolder
+from omegaconf import DictConfig, OmegaConf
 from pydantic.networks import AnyHttpUrl
 from requests import Session
 
@@ -21,28 +21,30 @@ from invokeai.app.services.download import DownloadJob, DownloadQueueServiceBase
 from invokeai.app.services.events.events_base import EventServiceBase
 from invokeai.app.services.invoker import Invoker
 from invokeai.app.services.model_records import DuplicateModelException, ModelRecordServiceBase
+from invokeai.app.services.model_records.model_records_base import ModelRecordChanges
 from invokeai.backend.model_manager.config import (
     AnyModelConfig,
     BaseModelType,
+    CheckpointConfigBase,
     InvalidModelConfigException,
     ModelRepoVariant,
+    ModelSourceType,
     ModelType,
 )
-from invokeai.backend.model_manager.hash import FastModelHash
 from invokeai.backend.model_manager.metadata import (
     AnyModelRepoMetadata,
-    CivitaiMetadataFetch,
     HuggingFaceMetadataFetch,
     ModelMetadataWithFiles,
     RemoteModelFile,
 )
+from invokeai.backend.model_manager.metadata.metadata_base import HuggingFaceMetadata
 from invokeai.backend.model_manager.probe import ModelProbe
 from invokeai.backend.model_manager.search import ModelSearch
 from invokeai.backend.util import Chdir, InvokeAILogger
 from invokeai.backend.util.devices import choose_precision, choose_torch_device
 
 from .model_install_base import (
-    CivitaiModelSource,
+    MODEL_SOURCE_TO_TYPE_MAP,
     HFModelSource,
     InstallStatus,
     LocalModelSource,
@@ -91,7 +93,6 @@ class ModelInstallService(ModelInstallServiceBase):
         self._running = False
         self._session = session
         self._next_job_id = 0
-        self._metadata_store = record_store.metadata_store  # for convenience
 
     @property
     def app_config(self) -> InvokeAIAppConfig:  # noqa D102
@@ -114,6 +115,7 @@ class ModelInstallService(ModelInstallServiceBase):
                 raise Exception("Attempt to start the installer service twice")
             self._start_installer_thread()
             self._remove_dangling_install_dirs()
+            self._migrate_yaml()
             self.sync_to_config()
 
     def stop(self, invoker: Optional[Invoker] = None) -> None:
@@ -131,6 +133,14 @@ class ModelInstallService(ModelInstallServiceBase):
             self._download_cache.clear()
             self._running = False
 
+    def _put_in_queue(self, job: ModelInstallJob) -> None:
+        print(f'DEBUG: in _put_in_queue(job={job.id})')
+        if self._stop_event.is_set():
+            self.cancel_job(job)
+        else:
+            print(f'DEBUG: putting {job.id} into the install queue')
+            self._install_queue.put(job)
+
     def register_path(
         self,
         model_path: Union[Path, str],
@@ -140,6 +150,7 @@ class ModelInstallService(ModelInstallServiceBase):
         config = config or {}
         if not config.get("source"):
             config["source"] = model_path.resolve().as_posix()
+        config["source_type"] = ModelSourceType.Path
         return self._register(model_path, config)
 
     def install_path(
@@ -149,11 +160,8 @@ class ModelInstallService(ModelInstallServiceBase):
     ) -> str:  # noqa D102
         model_path = Path(model_path)
         config = config or {}
-        if not config.get("source"):
-            config["source"] = model_path.resolve().as_posix()
 
-        info: AnyModelConfig = self._probe_model(Path(model_path), config)
-        old_hash = info.current_hash
+        info: AnyModelConfig = ModelProbe.probe(Path(model_path), config, hash_algo=self._app_config.hashing_algorithm)
 
         if preferred_name := config.get("name"):
             preferred_name = Path(preferred_name).with_suffix(model_path.suffix)
@@ -167,8 +175,6 @@ class ModelInstallService(ModelInstallServiceBase):
             raise DuplicateModelException(
                 f"A model named {model_path.name} is already installed at {dest_path.as_posix()}"
             ) from excp
-        new_hash = FastModelHash.hash(new_path)
-        assert new_hash == old_hash, f"{model_path}: Model hash changed during installation, possibly corrupted."
 
         return self._register(
             new_path,
@@ -181,13 +187,14 @@ class ModelInstallService(ModelInstallServiceBase):
         source: str,
         config: Optional[Dict[str, Any]] = None,
         access_token: Optional[str] = None,
+        inplace: Optional[bool] = False,
     ) -> ModelInstallJob:
         variants = "|".join(ModelRepoVariant.__members__.values())
         hf_repoid_re = f"^([^/:]+/[^/:]+)(?::({variants})?(?::/?([^:]+))?)?$"
         source_obj: Optional[StringLikeSource] = None
 
         if Path(source).exists():  # A local file or directory
-            source_obj = LocalModelSource(path=Path(source))
+            source_obj = LocalModelSource(path=Path(source), inplace=inplace)
         elif match := re.match(hf_repoid_re, source):
             source_obj = HFModelSource(
                 repo_id=match.group(1),
@@ -196,9 +203,16 @@ class ModelInstallService(ModelInstallServiceBase):
                 access_token=access_token,
             )
         elif re.match(r"^https?://[^/]+", source):
+            # Pull the token from config if it exists and matches the URL
+            _token = access_token
+            if _token is None:
+                for pair in self.app_config.remote_api_tokens or []:
+                    if re.search(pair.url_regex, source):
+                        _token = pair.token
+                        break
             source_obj = URLModelSource(
                 url=AnyHttpUrl(source),
-                access_token=access_token,
+                access_token=_token,
             )
         else:
             raise ValueError(f"Unsupported model source: '{source}'")
@@ -212,9 +226,7 @@ class ModelInstallService(ModelInstallServiceBase):
 
         if isinstance(source, LocalModelSource):
             install_job = self._import_local_model(source, config)
-            self._install_queue.put(install_job)  # synchronously install
-        elif isinstance(source, CivitaiModelSource):
-            install_job = self._import_from_civitai(source, config)
+            self._put_in_queue(install_job)  # synchronously install
         elif isinstance(source, HFModelSource):
             install_job = self._import_from_hf(source, config)
         elif isinstance(source, URLModelSource):
@@ -281,8 +293,50 @@ class ModelInstallService(ModelInstallServiceBase):
             self._logger.info(f"{len(installed)} new models registered")
         self._logger.info("Model installer (re)initialized")
 
+    def _migrate_yaml(self) -> None:
+        db_models = self.record_store.all_models()
+        try:
+            yaml = self._get_yaml()
+        except OSError:
+            return
+
+        yaml_metadata = yaml.pop("__metadata__")
+        yaml_version = yaml_metadata.get("version")
+
+        if yaml_version != "3.0.0":
+            raise ValueError(
+                f"Attempted migration of unsupported `models.yaml` v{yaml_version}. Only v3.0.0 is supported. Exiting."
+            )
+
+        self._logger.info(
+            f"Starting one-time migration of {len(yaml.items())} models from `models.yaml` to database. This may take a few minutes."
+        )
+
+        if len(db_models) == 0 and len(yaml.items()) != 0:
+            for model_key, stanza in yaml.items():
+                _, _, model_name = str(model_key).split("/")
+                model_path = Path(stanza["path"])
+                if not model_path.is_absolute():
+                    model_path = self._app_config.models_path / model_path
+                model_path = model_path.resolve()
+
+                config: dict[str, Any] = {}
+                config["name"] = model_name
+                config["description"] = stanza.get("description")
+                config["config_path"] = stanza.get("config")
+
+                try:
+                    id = self.register_path(model_path=model_path, config=config)
+                    self._logger.info(f"Migrated {model_name} with id {id}")
+                except Exception as e:
+                    self._logger.warning(f"Model at {model_path} could not be migrated: {e}")
+
+        # Rename `models.yaml` to `models.yaml.bak` to prevent re-migration
+        yaml_path = self._app_config.model_conf_path
+        yaml_path.rename(yaml_path.with_suffix(".yaml.bak"))
+
     def scan_directory(self, scan_dir: Path, install: bool = False) -> List[str]:  # noqa D102
-        self._cached_model_paths = {Path(x.path) for x in self.record_store.all_models()}
+        self._cached_model_paths = {Path(x.path).resolve() for x in self.record_store.all_models()}
         callback = self._scan_install if install else self._scan_register
         search = ModelSearch(on_model_found=callback)
         self._models_installed.clear()
@@ -296,7 +350,7 @@ class ModelInstallService(ModelInstallServiceBase):
         """Unregister the model. Delete its files only if they are within our models directory."""
         model = self.record_store.get_model(key)
         models_dir = self.app_config.models_path
-        model_path = models_dir / model.path
+        model_path = models_dir / Path(model.path)  # handle legacy relative model paths
         if model_path.is_relative_to(models_dir):
             self.unconditionally_delete(key)
         else:
@@ -304,11 +358,11 @@ class ModelInstallService(ModelInstallServiceBase):
 
     def unconditionally_delete(self, key: str) -> None:  # noqa D102
         model = self.record_store.get_model(key)
-        path = self.app_config.models_path / model.path
-        if path.is_dir():
-            rmtree(path)
+        model_path = self.app_config.models_path / model.path
+        if model_path.is_dir():
+            rmtree(model_path)
         else:
-            path.unlink()
+            model_path.unlink()
         self.unregister(key)
 
     def download_and_cache(
@@ -357,10 +411,11 @@ class ModelInstallService(ModelInstallServiceBase):
                 done = True
                 continue
             try:
+                print(f'DEBUG: _install_next_item() checking for a job to install')
                 job = self._install_queue.get(timeout=1)
             except Empty:
                 continue
-
+            print(f'DEBUG: _install_next_item() got job {job.id}, status={job.status}')
             assert job.local_path is not None
             try:
                 if job.cancelled:
@@ -370,21 +425,23 @@ class ModelInstallService(ModelInstallServiceBase):
                     self._signal_job_errored(job)
 
                 elif (
-                    job.waiting or job.downloading
+                    job.waiting or job.downloads_done
                 ):  # local jobs will be in waiting state, remote jobs will be downloading state
                     job.total_bytes = self._stat_size(job.local_path)
                     job.bytes = job.total_bytes
                     self._signal_job_running(job)
                     job.config_in["source"] = str(job.source)
+                    job.config_in["source_type"] = MODEL_SOURCE_TO_TYPE_MAP[job.source.__class__]
+                    # enter the metadata, if there is any
+                    if isinstance(job.source_metadata, (HuggingFaceMetadata)):
+                        job.config_in["source_api_response"] = job.source_metadata.api_response
+
                     if job.inplace:
                         key = self.register_path(job.local_path, job.config_in)
                     else:
                         key = self.install_path(job.local_path, job.config_in)
                     job.config_out = self.record_store.get_model(key)
-
-                    # enter the metadata, if there is any
-                    if job.source_metadata:
-                        self._metadata_store.add_metadata(key, job.source_metadata)
+                    print(f'DEBUG: _install_next_item() signaling completion for job={job.id}, status={job.status}')
                     self._signal_job_completed(job)
 
             except InvalidModelConfigException as excp:
@@ -444,11 +501,13 @@ class ModelInstallService(ModelInstallServiceBase):
             self._logger.info(f"Scanning {self._app_config.models_path} for new and orphaned models")
             for cur_base_model in BaseModelType:
                 for cur_model_type in ModelType:
-                    models_dir = Path(cur_base_model.value, cur_model_type.value)
+                    models_dir = self._app_config.models_path / Path(cur_base_model.value, cur_model_type.value)
+                    if not models_dir.exists():
+                        continue
                     installed.update(self.scan_directory(models_dir))
             self._logger.info(f"{len(installed)} new models registered; {len(defunct_models)} unregistered")
 
-    def _sync_model_path(self, key: str, ignore_hash_change: bool = False) -> AnyModelConfig:
+    def _sync_model_path(self, key: str) -> AnyModelConfig:
         """
         Move model into the location indicated by its basetype, type and name.
 
@@ -463,21 +522,21 @@ class ModelInstallService(ModelInstallServiceBase):
         old_path = Path(model.path)
         models_dir = self.app_config.models_path
 
-        if not old_path.is_relative_to(models_dir):
+        try:
+            old_path.relative_to(models_dir)
+            return model
+        except ValueError:
+            pass
+
+        new_path = models_dir / model.base.value / model.type.value / old_path.name
+
+        if old_path == new_path or new_path.exists() and old_path == new_path.resolve():
             return model
 
-        new_path = models_dir / model.base.value / model.type.value / model.name
         self._logger.info(f"Moving {model.name} to {new_path}.")
         new_path = self._move_model(old_path, new_path)
-        new_hash = FastModelHash.hash(new_path)
-        model.path = new_path.relative_to(models_dir).as_posix()
-        if model.current_hash != new_hash:
-            assert (
-                ignore_hash_change
-            ), f"{model.name}: Model hash changed during installation, model is possibly corrupted"
-            model.current_hash = new_hash
-            self._logger.info(f"Model has new hash {model.current_hash}, but will continue to be identified by {key}")
-        self.record_store.update_model(key, model)
+        model.path = new_path.as_posix()
+        self.record_store.update_model(key, ModelRecordChanges(path=model.path))
         return model
 
     def _scan_register(self, model: Path) -> bool:
@@ -529,36 +588,22 @@ class ModelInstallService(ModelInstallServiceBase):
         move(old_path, new_path)
         return new_path
 
-    def _probe_model(self, model_path: Path, config: Optional[Dict[str, Any]] = None) -> AnyModelConfig:
-        info: AnyModelConfig = ModelProbe.probe(Path(model_path))
-        if config:  # used to override probe fields
-            for key, value in config.items():
-                setattr(info, key, value)
-        return info
-
-    def _create_key(self) -> str:
-        return sha256(randbytes(100)).hexdigest()[0:32]
-
     def _register(
         self, model_path: Path, config: Optional[Dict[str, Any]] = None, info: Optional[AnyModelConfig] = None
     ) -> str:
-        key = self._create_key()
-        if config and not config.get("key", None):
-            config["key"] = key
-        info = info or ModelProbe.probe(model_path, config)
+        config = config or {}
 
-        model_path = model_path.absolute()
-        if model_path.is_relative_to(self.app_config.models_path):
-            model_path = model_path.relative_to(self.app_config.models_path)
+        info = info or ModelProbe.probe(model_path, config, hash_algo=self._app_config.hashing_algorithm)
+
+        model_path = model_path.resolve()
 
         info.path = model_path.as_posix()
 
         # add 'main' specific fields
-        if hasattr(info, "config"):
-            # make config relative to our root
-            legacy_conf = (self.app_config.root_dir / self.app_config.legacy_conf_dir / info.config).resolve()
-            info.config = legacy_conf.relative_to(self.app_config.root_dir).as_posix()
-        self.record_store.add_model(info.key, info)
+        if isinstance(info, CheckpointConfigBase):
+            legacy_conf = (self.app_config.root_dir / self.app_config.legacy_conf_dir / info.config_path).resolve()
+            info.config_path = legacy_conf.as_posix()
+        self.record_store.add_model(info)
         return info.key
 
     def _next_id(self) -> int:
@@ -567,6 +612,16 @@ class ModelInstallService(ModelInstallServiceBase):
             self._next_job_id += 1
         return id
 
+    # --------------------------------------------------------------------------------------------
+    # Internal functions that manage the old yaml config
+    # --------------------------------------------------------------------------------------------
+    def _get_yaml(self) -> DictConfig:
+        """Fetch the models.yaml DictConfig for this installation."""
+        yaml_path = self._app_config.model_conf_path
+        omegaconf = OmegaConf.load(yaml_path)
+        assert isinstance(omegaconf, DictConfig)
+        return omegaconf
+
     @staticmethod
     def _guess_variant() -> Optional[ModelRepoVariant]:
         """Guess the best HuggingFace variant type to download."""
@@ -579,17 +634,9 @@ class ModelInstallService(ModelInstallServiceBase):
             source=source,
             config_in=config or {},
             local_path=Path(source.path),
-            inplace=source.inplace,
+            inplace=source.inplace or False,
         )
 
-    def _import_from_civitai(self, source: CivitaiModelSource, config: Optional[Dict[str, Any]]) -> ModelInstallJob:
-        if not source.access_token:
-            self._logger.info("No Civitai access token provided; some models may not be downloadable.")
-        metadata = CivitaiMetadataFetch(self._session).from_id(str(source.version_id))
-        assert isinstance(metadata, ModelMetadataWithFiles)
-        remote_files = metadata.download_urls(session=self._session)
-        return self._import_remote_model(source=source, config=config, metadata=metadata, remote_files=remote_files)
-
     def _import_from_hf(self, source: HFModelSource, config: Optional[Dict[str, Any]]) -> ModelInstallJob:
         # Add user's cached access token to HuggingFace requests
         source.access_token = source.access_token or HfFolder.get_token()
@@ -612,16 +659,16 @@ class ModelInstallService(ModelInstallServiceBase):
         )
 
     def _import_from_url(self, source: URLModelSource, config: Optional[Dict[str, Any]]) -> ModelInstallJob:
-        # URLs from Civitai or HuggingFace will be handled specially
-        url_patterns = {
-            r"^https?://civitai.com/": CivitaiMetadataFetch,
-            r"^https?://huggingface.co/[^/]+/[^/]+$": HuggingFaceMetadataFetch,
-        }
+        # URLs from HuggingFace will be handled specially
         metadata = None
-        for pattern, fetcher in url_patterns.items():
-            if re.match(pattern, str(source.url), re.IGNORECASE):
-                metadata = fetcher(self._session).from_url(source.url)
-                break
+        fetcher = None
+        try:
+            fetcher = self.get_fetcher_from_url(str(source.url))
+        except ValueError:
+            pass
+        kwargs: dict[str, Any] = {"session": self._session}
+        if fetcher is not None:
+            metadata = fetcher(**kwargs).from_url(source.url)
         self._logger.debug(f"metadata={metadata}")
         if metadata and isinstance(metadata, ModelMetadataWithFiles):
             remote_files = metadata.download_urls(session=self._session)
@@ -636,7 +683,7 @@ class ModelInstallService(ModelInstallServiceBase):
 
     def _import_remote_model(
         self,
-        source: ModelSource,
+        source: HFModelSource | URLModelSource,
         remote_files: List[RemoteModelFile],
         metadata: Optional[AnyModelRepoMetadata],
         config: Optional[Dict[str, Any]],
@@ -664,7 +711,7 @@ class ModelInstallService(ModelInstallServiceBase):
         # In the event that there is a subfolder specified in the source,
         # we need to remove it from the destination path in order to avoid
         # creating unwanted subfolders
-        if hasattr(source, "subfolder") and source.subfolder:
+        if isinstance(source, HFModelSource) and source.subfolder:
             root = Path(remote_files[0].path.parts[0])
             subfolder = root / source.subfolder
         else:
@@ -744,14 +791,16 @@ class ModelInstallService(ModelInstallServiceBase):
 
     def _download_complete_callback(self, download_job: DownloadJob) -> None:
         self._logger.info(f"{download_job.source}: model download complete")
+        print(f'DEBUG: _download_complete_callback(download_job={download_job.source}')
         with self._lock:
-            install_job = self._download_cache[download_job.source]
-            self._download_cache.pop(download_job.source, None)
-
+            install_job = self._download_cache.pop(download_job.source, None)
+            print(f'DEBUG: download_job={download_job.source} / install_job={install_job}')
             # are there any more active jobs left in this task?
-            if all(x.complete for x in install_job.download_parts):
-                #  now enqueue job for actual installation into the models directory
-                self._install_queue.put(install_job)
+            if install_job and install_job.downloading and all(x.complete for x in install_job.download_parts):
+                print(f'DEBUG: setting job {install_job.id} to DOWNLOADS_DONE')
+                install_job.status = InstallStatus.DOWNLOADS_DONE
+                print(f'DEBUG: putting {install_job.id} into the install queue')
+                self._put_in_queue(install_job)
 
             # Let other threads know that the number of downloads has changed
             self._downloads_changed_event.set()
@@ -793,7 +842,7 @@ class ModelInstallService(ModelInstallServiceBase):
 
         if all(x.in_terminal_state for x in install_job.download_parts):
             # When all parts have reached their terminal state, we finalize the job to clean up the temporary directory and other resources
-            self._install_queue.put(install_job)
+            self._put_in_queue(install_job)
 
     # ------------------------------------------------------------------------------------------------
     # Internal methods that put events on the event bus
@@ -850,4 +899,10 @@ class ModelInstallService(ModelInstallServiceBase):
     def _signal_job_cancelled(self, job: ModelInstallJob) -> None:
         self._logger.info(f"{job.source}: model installation was cancelled")
         if self._event_bus:
-            self._event_bus.emit_model_install_cancelled(str(job.source))
+            self._event_bus.emit_model_install_cancelled(str(job.source), id=job.id)
+
+    @staticmethod
+    def get_fetcher_from_url(url: str):
+        if re.match(r"^https?://huggingface.co/[^/]+/[^/]+$", url.lower()):
+            return HuggingFaceMetadataFetch
+        raise ValueError(f"Unsupported model source: '{url}'")

invokeai/app/services/model_load/model_load_default.py

@@ -68,6 +68,7 @@ class ModelLoadService(ModelLoadServiceBase):
             self._emit_load_event(
                 context_data=context_data,
                 model_config=model_config,
+                submodel_type=submodel_type,
             )
 
         implementation, model_config, submodel_type = self._registry.get_implementation(model_config, submodel_type)  # type: ignore
@@ -82,6 +83,7 @@ class ModelLoadService(ModelLoadServiceBase):
             self._emit_load_event(
                 context_data=context_data,
                 model_config=model_config,
+                submodel_type=submodel_type,
                 loaded=True,
             )
         return loaded_model
@@ -91,6 +93,7 @@ class ModelLoadService(ModelLoadServiceBase):
         context_data: InvocationContextData,
         model_config: AnyModelConfig,
         loaded: Optional[bool] = False,
+        submodel_type: Optional[SubModelType] = None,
     ) -> None:
         if not self._invoker:
             return
@@ -102,6 +105,7 @@ class ModelLoadService(ModelLoadServiceBase):
                 queue_batch_id=context_data.queue_item.batch_id,
                 graph_execution_state_id=context_data.queue_item.session_id,
                 model_config=model_config,
+                submodel_type=submodel_type,
             )
         else:
             self._invoker.services.events.emit_model_load_completed(
@@ -110,4 +114,5 @@ class ModelLoadService(ModelLoadServiceBase):
                 queue_batch_id=context_data.queue_item.batch_id,
                 graph_execution_state_id=context_data.queue_item.session_id,
                 model_config=model_config,
+                submodel_type=submodel_type,
             )

invokeai/app/services/model_manager/model_manager_base.py

@@ -1,15 +1,11 @@
 # Copyright (c) 2023 Lincoln D. Stein and the InvokeAI Team
 
 from abc import ABC, abstractmethod
-from typing import Optional
 
 import torch
 from typing_extensions import Self
 
 from invokeai.app.services.invoker import Invoker
-from invokeai.app.services.shared.invocation_context import InvocationContextData
-from invokeai.backend.model_manager.config import AnyModelConfig, BaseModelType, ModelType, SubModelType
-from invokeai.backend.model_manager.load.load_base import LoadedModel
 
 from ..config import InvokeAIAppConfig
 from ..download import DownloadQueueServiceBase
@@ -70,32 +66,3 @@ class ModelManagerServiceBase(ABC):
     @abstractmethod
     def stop(self, invoker: Invoker) -> None:
         pass
-
-    @abstractmethod
-    def load_model_by_config(
-        self,
-        model_config: AnyModelConfig,
-        submodel_type: Optional[SubModelType] = None,
-        context_data: Optional[InvocationContextData] = None,
-    ) -> LoadedModel:
-        pass
-
-    @abstractmethod
-    def load_model_by_key(
-        self,
-        key: str,
-        submodel_type: Optional[SubModelType] = None,
-        context_data: Optional[InvocationContextData] = None,
-    ) -> LoadedModel:
-        pass
-
-    @abstractmethod
-    def load_model_by_attr(
-        self,
-        model_name: str,
-        base_model: BaseModelType,
-        model_type: ModelType,
-        submodel: Optional[SubModelType] = None,
-        context_data: Optional[InvocationContextData] = None,
-    ) -> LoadedModel:
-        pass

invokeai/app/services/model_manager/model_manager_default.py

@@ -1,14 +1,10 @@
 # Copyright (c) 2023 Lincoln D. Stein and the InvokeAI Team
 """Implementation of ModelManagerServiceBase."""
 
-from typing import Optional
-
 import torch
 from typing_extensions import Self
 
 from invokeai.app.services.invoker import Invoker
-from invokeai.app.services.shared.invocation_context import InvocationContextData
-from invokeai.backend.model_manager import AnyModelConfig, BaseModelType, LoadedModel, ModelType, SubModelType
 from invokeai.backend.model_manager.load import ModelCache, ModelConvertCache, ModelLoaderRegistry
 from invokeai.backend.util.devices import choose_torch_device
 from invokeai.backend.util.logging import InvokeAILogger
@@ -18,7 +14,7 @@ from ..download import DownloadQueueServiceBase
 from ..events.events_base import EventServiceBase
 from ..model_install import ModelInstallService, ModelInstallServiceBase
 from ..model_load import ModelLoadService, ModelLoadServiceBase
-from ..model_records import ModelRecordServiceBase, UnknownModelException
+from ..model_records import ModelRecordServiceBase
 from .model_manager_base import ModelManagerServiceBase
 
 
@@ -64,56 +60,6 @@ class ModelManagerService(ModelManagerServiceBase):
             if hasattr(service, "stop"):
                 service.stop(invoker)
 
-    def load_model_by_config(
-        self,
-        model_config: AnyModelConfig,
-        submodel_type: Optional[SubModelType] = None,
-        context_data: Optional[InvocationContextData] = None,
-    ) -> LoadedModel:
-        return self.load.load_model(model_config, submodel_type, context_data)
-
-    def load_model_by_key(
-        self,
-        key: str,
-        submodel_type: Optional[SubModelType] = None,
-        context_data: Optional[InvocationContextData] = None,
-    ) -> LoadedModel:
-        config = self.store.get_model(key)
-        return self.load.load_model(config, submodel_type, context_data)
-
-    def load_model_by_attr(
-        self,
-        model_name: str,
-        base_model: BaseModelType,
-        model_type: ModelType,
-        submodel: Optional[SubModelType] = None,
-        context_data: Optional[InvocationContextData] = None,
-    ) -> LoadedModel:
-        """
-        Given a model's attributes, search the database for it, and if found, load and return the LoadedModel object.
-
-        This is provided for API compatability with the get_model() method
-        in the original model manager. However, note that LoadedModel is
-        not the same as the original ModelInfo that ws returned.
-
-        :param model_name: Name of to be fetched.
-        :param base_model: Base model
-        :param model_type: Type of the model
-        :param submodel: For main (pipeline models), the submodel to fetch
-        :param context: The invocation context.
-
-        Exceptions: UnknownModelException -- model with this key not known
-                    NotImplementedException -- a model loader was not provided at initialization time
-                    ValueError -- more than one model matches this combination
-        """
-        configs = self.store.search_by_attr(model_name, base_model, model_type)
-        if len(configs) == 0:
-            raise UnknownModelException(f"{base_model}/{model_type}/{model_name}: Unknown model")
-        elif len(configs) > 1:
-            raise ValueError(f"{base_model}/{model_type}/{model_name}: More than one model matches.")
-        else:
-            return self.load.load_model(configs[0], submodel, context_data)
-
     @classmethod
     def build_model_manager(
         cls,

invokeai/app/services/model_metadata/__init__.py

@@ -1,9 +0,0 @@
-"""Init file for ModelMetadataStoreService module."""
-
-from .metadata_store_base import ModelMetadataStoreBase
-from .metadata_store_sql import ModelMetadataStoreSQL
-
-__all__ = [
-    "ModelMetadataStoreBase",
-    "ModelMetadataStoreSQL",
-]

invokeai/app/services/model_metadata/metadata_store_base.py

@@ -1,65 +0,0 @@
-# Copyright (c) 2023 Lincoln D. Stein and the InvokeAI Development Team
-"""
-Storage for Model Metadata
-"""
-
-from abc import ABC, abstractmethod
-from typing import List, Set, Tuple
-
-from invokeai.backend.model_manager.metadata import AnyModelRepoMetadata
-
-
-class ModelMetadataStoreBase(ABC):
-    """Store, search and fetch model metadata retrieved from remote repositories."""
-
-    @abstractmethod
-    def add_metadata(self, model_key: str, metadata: AnyModelRepoMetadata) -> None:
-        """
-        Add a block of repo metadata to a model record.
-
-        The model record config must already exist in the database with the
-        same key. Otherwise a FOREIGN KEY constraint exception will be raised.
-
-        :param model_key: Existing model key in the `model_config` table
-        :param metadata: ModelRepoMetadata object to store
-        """
-
-    @abstractmethod
-    def get_metadata(self, model_key: str) -> AnyModelRepoMetadata:
-        """Retrieve the ModelRepoMetadata corresponding to model key."""
-
-    @abstractmethod
-    def list_all_metadata(self) -> List[Tuple[str, AnyModelRepoMetadata]]:  # key, metadata
-        """Dump out all the metadata."""
-
-    @abstractmethod
-    def update_metadata(self, model_key: str, metadata: AnyModelRepoMetadata) -> AnyModelRepoMetadata:
-        """
-        Update metadata corresponding to the model with the indicated key.
-
-        :param model_key: Existing model key in the `model_config` table
-        :param metadata: ModelRepoMetadata object to update
-        """
-
-    @abstractmethod
-    def list_tags(self) -> Set[str]:
-        """Return all tags in the tags table."""
-
-    @abstractmethod
-    def search_by_tag(self, tags: Set[str]) -> Set[str]:
-        """Return the keys of models containing all of the listed tags."""
-
-    @abstractmethod
-    def search_by_author(self, author: str) -> Set[str]:
-        """Return the keys of models authored by the indicated author."""
-
-    @abstractmethod
-    def search_by_name(self, name: str) -> Set[str]:
-        """
-        Return the keys of models with the indicated name.
-
-        Note that this is the name of the model given to it by
-        the remote source. The user may have changed the local
-        name. The local name will be located in the model config
-        record object.
-        """

invokeai/app/services/model_metadata/metadata_store_sql.py

@@ -1,222 +0,0 @@
-# Copyright (c) 2023 Lincoln D. Stein and the InvokeAI Development Team
-"""
-SQL Storage for Model Metadata
-"""
-
-import sqlite3
-from typing import List, Optional, Set, Tuple
-
-from invokeai.app.services.shared.sqlite.sqlite_database import SqliteDatabase
-from invokeai.backend.model_manager.metadata import AnyModelRepoMetadata, UnknownMetadataException
-from invokeai.backend.model_manager.metadata.fetch import ModelMetadataFetchBase
-
-from .metadata_store_base import ModelMetadataStoreBase
-
-
-class ModelMetadataStoreSQL(ModelMetadataStoreBase):
-    """Store, search and fetch model metadata retrieved from remote repositories."""
-
-    def __init__(self, db: SqliteDatabase):
-        """
-        Initialize a new object from preexisting sqlite3 connection and threading lock objects.
-
-        :param conn: sqlite3 connection object
-        :param lock: threading Lock object
-        """
-        super().__init__()
-        self._db = db
-        self._cursor = self._db.conn.cursor()
-
-    def add_metadata(self, model_key: str, metadata: AnyModelRepoMetadata) -> None:
-        """
-        Add a block of repo metadata to a model record.
-
-        The model record config must already exist in the database with the
-        same key. Otherwise a FOREIGN KEY constraint exception will be raised.
-
-        :param model_key: Existing model key in the `model_config` table
-        :param metadata: ModelRepoMetadata object to store
-        """
-        json_serialized = metadata.model_dump_json()
-        with self._db.lock:
-            try:
-                self._cursor.execute(
-                    """--sql
-                    INSERT INTO model_metadata(
-                       id,
-                       metadata
-                    )
-                    VALUES (?,?);
-                    """,
-                    (
-                        model_key,
-                        json_serialized,
-                    ),
-                )
-                self._update_tags(model_key, metadata.tags)
-                self._db.conn.commit()
-            except sqlite3.IntegrityError as excp:  # FOREIGN KEY error: the key was not in model_config table
-                self._db.conn.rollback()
-                raise UnknownMetadataException from excp
-            except sqlite3.Error as excp:
-                self._db.conn.rollback()
-                raise excp
-
-    def get_metadata(self, model_key: str) -> AnyModelRepoMetadata:
-        """Retrieve the ModelRepoMetadata corresponding to model key."""
-        with self._db.lock:
-            self._cursor.execute(
-                """--sql
-                SELECT metadata FROM model_metadata
-                WHERE id=?;
-                """,
-                (model_key,),
-            )
-            rows = self._cursor.fetchone()
-            if not rows:
-                raise UnknownMetadataException("model metadata not found")
-            return ModelMetadataFetchBase.from_json(rows[0])
-
-    def list_all_metadata(self) -> List[Tuple[str, AnyModelRepoMetadata]]:  # key, metadata
-        """Dump out all the metadata."""
-        with self._db.lock:
-            self._cursor.execute(
-                """--sql
-                SELECT id,metadata FROM model_metadata;
-                """,
-                (),
-            )
-            rows = self._cursor.fetchall()
-        return [(x[0], ModelMetadataFetchBase.from_json(x[1])) for x in rows]
-
-    def update_metadata(self, model_key: str, metadata: AnyModelRepoMetadata) -> AnyModelRepoMetadata:
-        """
-        Update metadata corresponding to the model with the indicated key.
-
-        :param model_key: Existing model key in the `model_config` table
-        :param metadata: ModelRepoMetadata object to update
-        """
-        json_serialized = metadata.model_dump_json()  # turn it into a json string.
-        with self._db.lock:
-            try:
-                self._cursor.execute(
-                    """--sql
-                    UPDATE model_metadata
-                    SET
-                        metadata=?
-                    WHERE id=?;
-                    """,
-                    (json_serialized, model_key),
-                )
-                if self._cursor.rowcount == 0:
-                    raise UnknownMetadataException("model metadata not found")
-                self._update_tags(model_key, metadata.tags)
-                self._db.conn.commit()
-            except sqlite3.Error as e:
-                self._db.conn.rollback()
-                raise e
-
-        return self.get_metadata(model_key)
-
-    def list_tags(self) -> Set[str]:
-        """Return all tags in the tags table."""
-        self._cursor.execute(
-            """--sql
-            select tag_text from tags;
-            """
-        )
-        return {x[0] for x in self._cursor.fetchall()}
-
-    def search_by_tag(self, tags: Set[str]) -> Set[str]:
-        """Return the keys of models containing all of the listed tags."""
-        with self._db.lock:
-            try:
-                matches: Optional[Set[str]] = None
-                for tag in tags:
-                    self._cursor.execute(
-                        """--sql
-                        SELECT a.model_id FROM model_tags AS a,
-                                                     tags AS b
-                        WHERE a.tag_id=b.tag_id
-                          AND b.tag_text=?;
-                        """,
-                        (tag,),
-                    )
-                    model_keys = {x[0] for x in self._cursor.fetchall()}
-                    if matches is None:
-                        matches = model_keys
-                    matches = matches.intersection(model_keys)
-            except sqlite3.Error as e:
-                raise e
-        return matches if matches else set()
-
-    def search_by_author(self, author: str) -> Set[str]:
-        """Return the keys of models authored by the indicated author."""
-        self._cursor.execute(
-            """--sql
-            SELECT id FROM model_metadata
-            WHERE author=?;
-            """,
-            (author,),
-        )
-        return {x[0] for x in self._cursor.fetchall()}
-
-    def search_by_name(self, name: str) -> Set[str]:
-        """
-        Return the keys of models with the indicated name.
-
-        Note that this is the name of the model given to it by
-        the remote source. The user may have changed the local
-        name. The local name will be located in the model config
-        record object.
-        """
-        self._cursor.execute(
-            """--sql
-            SELECT id FROM model_metadata
-            WHERE name=?;
-            """,
-            (name,),
-        )
-        return {x[0] for x in self._cursor.fetchall()}
-
-    def _update_tags(self, model_key: str, tags: Set[str]) -> None:
-        """Update tags for the model referenced by model_key."""
-        # remove previous tags from this model
-        self._cursor.execute(
-            """--sql
-            DELETE FROM model_tags
-            WHERE model_id=?;
-            """,
-            (model_key,),
-        )
-
-        for tag in tags:
-            self._cursor.execute(
-                """--sql
-                INSERT OR IGNORE INTO tags (
-                  tag_text
-                  )
-                VALUES (?);
-                """,
-                (tag,),
-            )
-            self._cursor.execute(
-                """--sql
-                SELECT tag_id
-                FROM tags
-                WHERE tag_text = ?
-                LIMIT 1;
-                """,
-                (tag,),
-            )
-            tag_id = self._cursor.fetchone()[0]
-            self._cursor.execute(
-                """--sql
-                INSERT OR IGNORE INTO model_tags (
-                   model_id,
-                   tag_id
-                  )
-                VALUES (?,?);
-                """,
-                (model_key, tag_id),
-            )

invokeai/app/services/model_records/__init__.py

@@ -1,4 +1,5 @@
 """Init file for model record services."""
+
 from .model_records_base import (  # noqa F401
     DuplicateModelException,
     InvalidModelException,

invokeai/app/services/model_records/model_records_base.py

@@ -6,20 +6,24 @@ Abstract base class for storing and retrieving model configuration records.
 from abc import ABC, abstractmethod
 from enum import Enum
 from pathlib import Path
-from typing import Any, Dict, List, Optional, Set, Tuple, Union
+from typing import List, Optional, Set, Union
 
 from pydantic import BaseModel, Field
 
 from invokeai.app.services.shared.pagination import PaginatedResults
+from invokeai.app.util.model_exclude_null import BaseModelExcludeNull
 from invokeai.backend.model_manager import (
     AnyModelConfig,
     BaseModelType,
     ModelFormat,
     ModelType,
 )
-from invokeai.backend.model_manager.metadata import AnyModelRepoMetadata
-
-from ..model_metadata import ModelMetadataStoreBase
+from invokeai.backend.model_manager.config import (
+    ControlAdapterDefaultSettings,
+    MainModelDefaultSettings,
+    ModelVariantType,
+    SchedulerPredictionType,
+)
 
 
 class DuplicateModelException(Exception):
@@ -60,11 +64,34 @@ class ModelSummary(BaseModel):
     tags: Set[str] = Field(description="tags associated with model")
 
 
+class ModelRecordChanges(BaseModelExcludeNull):
+    """A set of changes to apply to a model."""
+
+    # Changes applicable to all models
+    name: Optional[str] = Field(description="Name of the model.", default=None)
+    path: Optional[str] = Field(description="Path to the model.", default=None)
+    description: Optional[str] = Field(description="Model description", default=None)
+    base: Optional[BaseModelType] = Field(description="The base model.", default=None)
+    trigger_phrases: Optional[set[str]] = Field(description="Set of trigger phrases for this model", default=None)
+    default_settings: Optional[MainModelDefaultSettings | ControlAdapterDefaultSettings] = Field(
+        description="Default settings for this model", default=None
+    )
+
+    # Checkpoint-specific changes
+    # TODO(MM2): Should we expose these? Feels footgun-y...
+    variant: Optional[ModelVariantType] = Field(description="The variant of the model.", default=None)
+    prediction_type: Optional[SchedulerPredictionType] = Field(
+        description="The prediction type of the model.", default=None
+    )
+    upcast_attention: Optional[bool] = Field(description="Whether to upcast attention.", default=None)
+    config_path: Optional[str] = Field(description="Path to config file for model", default=None)
+
+
 class ModelRecordServiceBase(ABC):
     """Abstract base class for storage and retrieval of model configs."""
 
     @abstractmethod
-    def add_model(self, key: str, config: Union[Dict[str, Any], AnyModelConfig]) -> AnyModelConfig:
+    def add_model(self, config: AnyModelConfig) -> AnyModelConfig:
         """
         Add a model to the database.
 
@@ -88,13 +115,12 @@ class ModelRecordServiceBase(ABC):
         pass
 
     @abstractmethod
-    def update_model(self, key: str, config: Union[Dict[str, Any], AnyModelConfig]) -> AnyModelConfig:
+    def update_model(self, key: str, changes: ModelRecordChanges) -> AnyModelConfig:
         """
         Update the model, returning the updated version.
 
-        :param key: Unique key for the model to be updated
-        :param config: Model configuration record. Either a dict with the
-         required fields, or a ModelConfigBase instance.
+        :param key: Unique key for the model to be updated.
+        :param changes: A set of changes to apply to this model. Changes are validated before being written.
         """
         pass
 
@@ -109,40 +135,17 @@ class ModelRecordServiceBase(ABC):
         """
         pass
 
-    @property
     @abstractmethod
-    def metadata_store(self) -> ModelMetadataStoreBase:
-        """Return a ModelMetadataStore initialized on the same database."""
-        pass
-
-    @abstractmethod
-    def get_metadata(self, key: str) -> Optional[AnyModelRepoMetadata]:
+    def get_model_by_hash(self, hash: str) -> AnyModelConfig:
         """
-        Retrieve metadata (if any) from when model was downloaded from a repo.
+        Retrieve the configuration for the indicated model.
 
-        :param key: Model key
+        :param hash: Hash of model config to be fetched.
+
+        Exceptions: UnknownModelException
         """
         pass
 
-    @abstractmethod
-    def list_all_metadata(self) -> List[Tuple[str, AnyModelRepoMetadata]]:
-        """List metadata for all models that have it."""
-        pass
-
-    @abstractmethod
-    def search_by_metadata_tag(self, tags: Set[str]) -> List[AnyModelConfig]:
-        """
-        Search model metadata for ones with all listed tags and return their corresponding configs.
-
-        :param tags: Set of tags to search for. All tags must be present.
-        """
-        pass
-
-    @abstractmethod
-    def list_tags(self) -> Set[str]:
-        """Return a unique set of all the model tags in the metadata database."""
-        pass
-
     @abstractmethod
     def list_models(
         self, page: int = 0, per_page: int = 10, order_by: ModelRecordOrderBy = ModelRecordOrderBy.Default
@@ -217,21 +220,3 @@ class ModelRecordServiceBase(ABC):
                 f"More than one model matched the search criteria: base_model='{base_model}', model_type='{model_type}', model_name='{model_name}'."
             )
         return model_configs[0]
-
-    def rename_model(
-        self,
-        key: str,
-        new_name: str,
-    ) -> AnyModelConfig:
-        """
-        Rename the indicated model. Just a special case of update_model().
-
-        In some implementations, renaming the model may involve changing where
-        it is stored on the filesystem. So this is broken out.
-
-        :param key: Model key
-        :param new_name: New name for model
-        """
-        config = self.get_model(key)
-        config.name = new_name
-        return self.update_model(key, config)

invokeai/app/services/model_records/model_records_sql.py

@@ -39,12 +39,11 @@ Typical usage:
   configs = store.search_by_attr(base_model='sd-2', model_type='main')
 """
 
-
 import json
 import sqlite3
 from math import ceil
 from pathlib import Path
-from typing import Any, Dict, List, Optional, Set, Tuple, Union
+from typing import List, Optional, Union
 
 from invokeai.app.services.shared.pagination import PaginatedResults
 from invokeai.backend.model_manager.config import (
@@ -54,12 +53,11 @@ from invokeai.backend.model_manager.config import (
     ModelFormat,
     ModelType,
 )
-from invokeai.backend.model_manager.metadata import AnyModelRepoMetadata, UnknownMetadataException
 
-from ..model_metadata import ModelMetadataStoreBase, ModelMetadataStoreSQL
 from ..shared.sqlite.sqlite_database import SqliteDatabase
 from .model_records_base import (
     DuplicateModelException,
+    ModelRecordChanges,
     ModelRecordOrderBy,
     ModelRecordServiceBase,
     ModelSummary,
@@ -70,7 +68,7 @@ from .model_records_base import (
 class ModelRecordServiceSQL(ModelRecordServiceBase):
     """Implementation of the ModelConfigStore ABC using a SQL database."""
 
-    def __init__(self, db: SqliteDatabase, metadata_store: ModelMetadataStoreBase):
+    def __init__(self, db: SqliteDatabase):
         """
         Initialize a new object from preexisting sqlite3 connection and threading lock objects.
 
@@ -79,14 +77,13 @@ class ModelRecordServiceSQL(ModelRecordServiceBase):
         super().__init__()
         self._db = db
         self._cursor = db.conn.cursor()
-        self._metadata_store = metadata_store
 
     @property
     def db(self) -> SqliteDatabase:
         """Return the underlying database."""
         return self._db
 
-    def add_model(self, key: str, config: Union[Dict[str, Any], AnyModelConfig]) -> AnyModelConfig:
+    def add_model(self, config: AnyModelConfig) -> AnyModelConfig:
         """
         Add a model to the database.
 
@@ -96,23 +93,19 @@ class ModelRecordServiceSQL(ModelRecordServiceBase):
 
         Can raise DuplicateModelException and InvalidModelConfigException exceptions.
         """
-        record = ModelConfigFactory.make_config(config, key=key)  # ensure it is a valid config obect.
-        json_serialized = record.model_dump_json()  # and turn it into a json string.
         with self._db.lock:
             try:
                 self._cursor.execute(
                     """--sql
-                    INSERT INTO model_config (
+                    INSERT INTO models (
                        id,
-                       original_hash,
                        config
                       )
-                    VALUES (?,?,?);
+                    VALUES (?,?);
                     """,
                     (
-                        key,
-                        record.original_hash,
-                        json_serialized,
+                        config.key,
+                        config.model_dump_json(),
                     ),
                 )
                 self._db.conn.commit()
@@ -120,12 +113,12 @@ class ModelRecordServiceSQL(ModelRecordServiceBase):
             except sqlite3.IntegrityError as e:
                 self._db.conn.rollback()
                 if "UNIQUE constraint failed" in str(e):
-                    if "model_config.path" in str(e):
-                        msg = f"A model with path '{record.path}' is already installed"
-                    elif "model_config.name" in str(e):
-                        msg = f"A model with name='{record.name}', type='{record.type}', base='{record.base}' is already installed"
+                    if "models.path" in str(e):
+                        msg = f"A model with path '{config.path}' is already installed"
+                    elif "models.name" in str(e):
+                        msg = f"A model with name='{config.name}', type='{config.type}', base='{config.base}' is already installed"
                     else:
-                        msg = f"A model with key '{key}' is already installed"
+                        msg = f"A model with key '{config.key}' is already installed"
                     raise DuplicateModelException(msg) from e
                 else:
                     raise e
@@ -133,7 +126,7 @@ class ModelRecordServiceSQL(ModelRecordServiceBase):
                 self._db.conn.rollback()
                 raise e
 
-        return self.get_model(key)
+        return self.get_model(config.key)
 
     def del_model(self, key: str) -> None:
         """
@@ -147,7 +140,7 @@ class ModelRecordServiceSQL(ModelRecordServiceBase):
             try:
                 self._cursor.execute(
                     """--sql
-                    DELETE FROM model_config
+                    DELETE FROM models
                     WHERE id=?;
                     """,
                     (key,),
@@ -159,21 +152,20 @@ class ModelRecordServiceSQL(ModelRecordServiceBase):
                 self._db.conn.rollback()
                 raise e
 
-    def update_model(self, key: str, config: Union[Dict[str, Any], AnyModelConfig]) -> AnyModelConfig:
-        """
-        Update the model, returning the updated version.
+    def update_model(self, key: str, changes: ModelRecordChanges) -> AnyModelConfig:
+        record = self.get_model(key)
+
+        # Model configs use pydantic's `validate_assignment`, so each change is validated by pydantic.
+        for field_name in changes.model_fields_set:
+            setattr(record, field_name, getattr(changes, field_name))
+
+        json_serialized = record.model_dump_json()
 
-        :param key: Unique key for the model to be updated
-        :param config: Model configuration record. Either a dict with the
-         required fields, or a ModelConfigBase instance.
-        """
-        record = ModelConfigFactory.make_config(config, key=key)  # ensure it is a valid config obect
-        json_serialized = record.model_dump_json()  # and turn it into a json string.
         with self._db.lock:
             try:
                 self._cursor.execute(
                     """--sql
-                    UPDATE model_config
+                    UPDATE models
                     SET
                         config=?
                     WHERE id=?;
@@ -200,7 +192,7 @@ class ModelRecordServiceSQL(ModelRecordServiceBase):
         with self._db.lock:
             self._cursor.execute(
                 """--sql
-                SELECT config, strftime('%s',updated_at) FROM model_config
+                SELECT config, strftime('%s',updated_at) FROM models
                 WHERE id=?;
                 """,
                 (key,),
@@ -211,6 +203,21 @@ class ModelRecordServiceSQL(ModelRecordServiceBase):
             model = ModelConfigFactory.make_config(json.loads(rows[0]), timestamp=rows[1])
         return model
 
+    def get_model_by_hash(self, hash: str) -> AnyModelConfig:
+        with self._db.lock:
+            self._cursor.execute(
+                """--sql
+                SELECT config, strftime('%s',updated_at) FROM models
+                WHERE hash=?;
+                """,
+                (hash,),
+            )
+            rows = self._cursor.fetchone()
+            if not rows:
+                raise UnknownModelException("model not found")
+            model = ModelConfigFactory.make_config(json.loads(rows[0]), timestamp=rows[1])
+        return model
+
     def exists(self, key: str) -> bool:
         """
         Return True if a model with the indicated key exists in the databse.
@@ -221,7 +228,7 @@ class ModelRecordServiceSQL(ModelRecordServiceBase):
         with self._db.lock:
             self._cursor.execute(
                 """--sql
-                select count(*) FROM model_config
+                select count(*) FROM models
                 WHERE id=?;
                 """,
                 (key,),
@@ -235,6 +242,7 @@ class ModelRecordServiceSQL(ModelRecordServiceBase):
         base_model: Optional[BaseModelType] = None,
         model_type: Optional[ModelType] = None,
         model_format: Optional[ModelFormat] = None,
+        order_by: ModelRecordOrderBy = ModelRecordOrderBy.Default,
     ) -> List[AnyModelConfig]:
         """
         Return models matching name, base and/or type.
@@ -243,13 +251,23 @@ class ModelRecordServiceSQL(ModelRecordServiceBase):
         :param base_model: Filter by base model (optional)
         :param model_type: Filter by type of model (optional)
         :param model_format: Filter by model format (e.g. "diffusers") (optional)
+        :param order_by: Result order
 
         If none of the optional filters are passed, will return all
         models in the database.
         """
-        results = []
-        where_clause = []
-        bindings = []
+
+        assert isinstance(order_by, ModelRecordOrderBy)
+        ordering = {
+            ModelRecordOrderBy.Default: "type, base, name, format",
+            ModelRecordOrderBy.Type: "type",
+            ModelRecordOrderBy.Base: "base",
+            ModelRecordOrderBy.Name: "name",
+            ModelRecordOrderBy.Format: "format",
+        }
+
+        where_clause: list[str] = []
+        bindings: list[str] = []
         if model_name:
             where_clause.append("name=?")
             bindings.append(model_name)
@@ -266,14 +284,15 @@ class ModelRecordServiceSQL(ModelRecordServiceBase):
         with self._db.lock:
             self._cursor.execute(
                 f"""--sql
-                select config, strftime('%s',updated_at) FROM model_config
-                {where};
+                SELECT config, strftime('%s',updated_at)
+                FROM models
+                {where}
+                ORDER BY {ordering[order_by]} -- using ? to bind doesn't work here for some reason;
                 """,
                 tuple(bindings),
             )
-            results = [
-                ModelConfigFactory.make_config(json.loads(x[0]), timestamp=x[1]) for x in self._cursor.fetchall()
-            ]
+            result = self._cursor.fetchall()
+            results = [ModelConfigFactory.make_config(json.loads(x[0]), timestamp=x[1]) for x in result]
         return results
 
     def search_by_path(self, path: Union[str, Path]) -> List[AnyModelConfig]:
@@ -282,7 +301,7 @@ class ModelRecordServiceSQL(ModelRecordServiceBase):
         with self._db.lock:
             self._cursor.execute(
                 """--sql
-                SELECT config, strftime('%s',updated_at) FROM model_config
+                SELECT config, strftime('%s',updated_at) FROM models
                 WHERE path=?;
                 """,
                 (str(path),),
@@ -293,13 +312,13 @@ class ModelRecordServiceSQL(ModelRecordServiceBase):
         return results
 
     def search_by_hash(self, hash: str) -> List[AnyModelConfig]:
-        """Return models with the indicated original_hash."""
+        """Return models with the indicated hash."""
         results = []
         with self._db.lock:
             self._cursor.execute(
                 """--sql
-                SELECT config, strftime('%s',updated_at) FROM model_config
-                WHERE original_hash=?;
+                SELECT config, strftime('%s',updated_at) FROM models
+                WHERE hash=?;
                 """,
                 (hash,),
             )
@@ -308,83 +327,35 @@ class ModelRecordServiceSQL(ModelRecordServiceBase):
             ]
         return results
 
-    @property
-    def metadata_store(self) -> ModelMetadataStoreBase:
-        """Return a ModelMetadataStore initialized on the same database."""
-        return self._metadata_store
-
-    def get_metadata(self, key: str) -> Optional[AnyModelRepoMetadata]:
-        """
-        Retrieve metadata (if any) from when model was downloaded from a repo.
-
-        :param key: Model key
-        """
-        store = self.metadata_store
-        try:
-            metadata = store.get_metadata(key)
-            return metadata
-        except UnknownMetadataException:
-            return None
-
-    def search_by_metadata_tag(self, tags: Set[str]) -> List[AnyModelConfig]:
-        """
-        Search model metadata for ones with all listed tags and return their corresponding configs.
-
-        :param tags: Set of tags to search for. All tags must be present.
-        """
-        store = ModelMetadataStoreSQL(self._db)
-        keys = store.search_by_tag(tags)
-        return [self.get_model(x) for x in keys]
-
-    def list_tags(self) -> Set[str]:
-        """Return a unique set of all the model tags in the metadata database."""
-        store = ModelMetadataStoreSQL(self._db)
-        return store.list_tags()
-
-    def list_all_metadata(self) -> List[Tuple[str, AnyModelRepoMetadata]]:
-        """List metadata for all models that have it."""
-        store = ModelMetadataStoreSQL(self._db)
-        return store.list_all_metadata()
-
     def list_models(
         self, page: int = 0, per_page: int = 10, order_by: ModelRecordOrderBy = ModelRecordOrderBy.Default
     ) -> PaginatedResults[ModelSummary]:
         """Return a paginated summary listing of each model in the database."""
+        assert isinstance(order_by, ModelRecordOrderBy)
         ordering = {
-            ModelRecordOrderBy.Default: "a.type, a.base, a.format, a.name",
-            ModelRecordOrderBy.Type: "a.type",
-            ModelRecordOrderBy.Base: "a.base",
-            ModelRecordOrderBy.Name: "a.name",
-            ModelRecordOrderBy.Format: "a.format",
+            ModelRecordOrderBy.Default: "type, base, name, format",
+            ModelRecordOrderBy.Type: "type",
+            ModelRecordOrderBy.Base: "base",
+            ModelRecordOrderBy.Name: "name",
+            ModelRecordOrderBy.Format: "format",
         }
 
-        def _fixup(summary: Dict[str, str]) -> Dict[str, Union[str, int, Set[str]]]:
-            """Fix up results so that there are no null values."""
-            result: Dict[str, Union[str, int, Set[str]]] = {}
-            for key, item in summary.items():
-                result[key] = item or ""
-            result["tags"] = set(json.loads(summary["tags"] or "[]"))
-            return result
-
         # Lock so that the database isn't updated while we're doing the two queries.
         with self._db.lock:
             # query1: get the total number of model configs
             self._cursor.execute(
                 """--sql
-                select count(*) from model_config;
+                select count(*) from models;
                 """,
                 (),
             )
             total = int(self._cursor.fetchone()[0])
 
-            # query2: fetch key fields from the join of model_config and model_metadata
+            # query2: fetch key fields
             self._cursor.execute(
                 f"""--sql
-                SELECT a.id as key, a.type, a.base, a.format, a.name,
-                       json_extract(a.config, '$.description') as description,
-                       json_extract(b.metadata, '$.tags') as tags
-                FROM model_config AS a
-                LEFT JOIN model_metadata AS b on a.id=b.id
+                SELECT config
+                FROM models
                 ORDER BY {ordering[order_by]} -- using ? to bind doesn't work here for some reason
                 LIMIT ?
                 OFFSET ?;
@@ -395,7 +366,7 @@ class ModelRecordServiceSQL(ModelRecordServiceBase):
                 ),
             )
             rows = self._cursor.fetchall()
-            items = [ModelSummary.model_validate(_fixup(dict(x))) for x in rows]
+            items = [ModelSummary.model_validate(dict(x)) for x in rows]
             return PaginatedResults(
                 page=page, pages=ceil(total / per_page), per_page=per_page, total=total, items=items
             )

invokeai/app/services/session_processor/session_processor_default.py

@@ -200,6 +200,7 @@ class DefaultSessionProcessor(SessionProcessorBase):
                                 self._invoker.services.logger.error(
                                     f"Error while invoking session {self._queue_item.session_id}, invocation {self._invocation.id} ({self._invocation.get_type()}):\n{e}"
                                 )
+                                self._invoker.services.logger.error(error)
 
                                 # Send error event
                                 self._invoker.services.events.emit_invocation_error(

invokeai/app/services/shared/invocation_context.py

@@ -1,7 +1,7 @@
 import threading
 from dataclasses import dataclass
 from pathlib import Path
-from typing import TYPE_CHECKING, Optional
+from typing import TYPE_CHECKING, Optional, Union
 
 from PIL.Image import Image
 from torch import Tensor
@@ -13,15 +13,16 @@ from invokeai.app.services.config.config_default import InvokeAIAppConfig
 from invokeai.app.services.image_records.image_records_common import ImageCategory, ResourceOrigin
 from invokeai.app.services.images.images_common import ImageDTO
 from invokeai.app.services.invocation_services import InvocationServices
+from invokeai.app.services.model_records.model_records_base import UnknownModelException
 from invokeai.app.util.step_callback import stable_diffusion_step_callback
 from invokeai.backend.model_manager.config import AnyModelConfig, BaseModelType, ModelFormat, ModelType, SubModelType
 from invokeai.backend.model_manager.load.load_base import LoadedModel
-from invokeai.backend.model_manager.metadata.metadata_base import AnyModelRepoMetadata
 from invokeai.backend.stable_diffusion.diffusers_pipeline import PipelineIntermediateState
 from invokeai.backend.stable_diffusion.diffusion.conditioning_data import ConditioningFieldData
 
 if TYPE_CHECKING:
     from invokeai.app.invocations.baseinvocation import BaseInvocation
+    from invokeai.app.invocations.model import ModelIdentifierField
     from invokeai.app.services.session_queue.session_queue_common import SessionQueueItem
 
 """
@@ -65,75 +66,86 @@ class InvocationContextInterface:
 
 class BoardsInterface(InvocationContextInterface):
     def create(self, board_name: str) -> BoardDTO:
-        """
-        Creates a board.
+        """Creates a board.
 
-        :param board_name: The name of the board to create.
+        Args:
+            board_name: The name of the board to create.
+
+        Returns:
+            The created board DTO.
         """
         return self._services.boards.create(board_name)
 
     def get_dto(self, board_id: str) -> BoardDTO:
-        """
-        Gets a board DTO.
+        """Gets a board DTO.
 
-        :param board_id: The ID of the board to get.
+        Args:
+            board_id: The ID of the board to get.
+
+        Returns:
+            The board DTO.
         """
         return self._services.boards.get_dto(board_id)
 
     def get_all(self) -> list[BoardDTO]:
-        """
-        Gets all boards.
+        """Gets all boards.
+
+        Returns:
+            A list of all boards.
         """
         return self._services.boards.get_all()
 
     def add_image_to_board(self, board_id: str, image_name: str) -> None:
-        """
-        Adds an image to a board.
+        """Adds an image to a board.
 
-        :param board_id: The ID of the board to add the image to.
-        :param image_name: The name of the image to add to the board.
+        Args:
+            board_id: The ID of the board to add the image to.
+            image_name: The name of the image to add to the board.
         """
         return self._services.board_images.add_image_to_board(board_id, image_name)
 
     def get_all_image_names_for_board(self, board_id: str) -> list[str]:
-        """
-        Gets all image names for a board.
+        """Gets all image names for a board.
 
-        :param board_id: The ID of the board to get the image names for.
+        Args:
+            board_id: The ID of the board to get the image names for.
+
+        Returns:
+            A list of all image names for the board.
         """
         return self._services.board_images.get_all_board_image_names_for_board(board_id)
 
 
 class LoggerInterface(InvocationContextInterface):
     def debug(self, message: str) -> None:
-        """
-        Logs a debug message.
+        """Logs a debug message.
 
-        :param message: The message to log.
+        Args:
+            message: The message to log.
         """
         self._services.logger.debug(message)
 
     def info(self, message: str) -> None:
-        """
-        Logs an info message.
+        """Logs an info message.
 
-        :param message: The message to log.
+        Args:
+            message: The message to log.
         """
         self._services.logger.info(message)
 
     def warning(self, message: str) -> None:
-        """
-        Logs a warning message.
+        """Logs a warning message.
 
-        :param message: The message to log.
+        Args:
+            message: The message to log.
         """
         self._services.logger.warning(message)
 
     def error(self, message: str) -> None:
-        """
-        Logs an error message.
+        """Logs an error message.
 
-        :param message: The message to log.
+        Args:
+            message: The message to log.
         """
         self._services.logger.error(message)
 
@@ -146,20 +158,23 @@ class ImagesInterface(InvocationContextInterface):
         image_category: ImageCategory = ImageCategory.GENERAL,
         metadata: Optional[MetadataField] = None,
     ) -> ImageDTO:
-        """
-        Saves an image, returning its DTO.
+        """Saves an image, returning its DTO.
 
         If the current queue item has a workflow or metadata, it is automatically saved with the image.
 
-        :param image: The image to save, as a PIL image.
-        :param board_id: The board ID to add the image to, if it should be added. It the invocation \
+        Args:
+            image: The image to save, as a PIL image.
+            board_id: The board ID to add the image to, if it should be added. It the invocation \
             inherits from `WithBoard`, that board will be used automatically. **Use this only if \
             you want to override or provide a board manually!**
-        :param image_category: The category of the image. Only the GENERAL category is added \
+            image_category: The category of the image. Only the GENERAL category is added \
             to the gallery.
-        :param metadata: The metadata to save with the image, if it should have any. If the \
+            metadata: The metadata to save with the image, if it should have any. If the \
             invocation inherits from `WithMetadata`, that metadata will be used automatically. \
             **Use this only if you want to override or provide metadata manually!**
+
+        Returns:
+            The saved image DTO.
         """
 
         # If `metadata` is provided directly, use that. Else, use the metadata provided by `WithMetadata`, falling back to None.
@@ -189,11 +204,14 @@ class ImagesInterface(InvocationContextInterface):
         )
 
     def get_pil(self, image_name: str, mode: IMAGE_MODES | None = None) -> Image:
-        """
-        Gets an image as a PIL Image object.
+        """Gets an image as a PIL Image object.
 
-        :param image_name: The name of the image to get.
-        :param mode: The color mode to convert the image to. If None, the original mode is used.
+        Args:
+            image_name: The name of the image to get.
+            mode: The color mode to convert the image to. If None, the original mode is used.
+
+        Returns:
+            The image as a PIL Image object.
         """
         image = self._services.images.get_pil_image(image_name)
         if mode and mode != image.mode:
@@ -206,158 +224,204 @@ class ImagesInterface(InvocationContextInterface):
         return image
 
     def get_metadata(self, image_name: str) -> Optional[MetadataField]:
-        """
-        Gets an image's metadata, if it has any.
+        """Gets an image's metadata, if it has any.
 
-        :param image_name: The name of the image to get the metadata for.
+        Args:
+            image_name: The name of the image to get the metadata for.
+
+        Returns:
+            The image's metadata, if it has any.
         """
         return self._services.images.get_metadata(image_name)
 
     def get_dto(self, image_name: str) -> ImageDTO:
-        """
-        Gets an image as an ImageDTO object.
+        """Gets an image as an ImageDTO object.
 
-        :param image_name: The name of the image to get.
+        Args:
+            image_name: The name of the image to get.
+
+        Returns:
+            The image as an ImageDTO object.
         """
         return self._services.images.get_dto(image_name)
 
 
 class TensorsInterface(InvocationContextInterface):
     def save(self, tensor: Tensor) -> str:
-        """
-        Saves a tensor, returning its name.
+        """Saves a tensor, returning its name.
 
-        :param tensor: The tensor to save.
+        Args:
+            tensor: The tensor to save.
+
+        Returns:
+            The name of the saved tensor.
         """
 
         name = self._services.tensors.save(obj=tensor)
         return name
 
     def load(self, name: str) -> Tensor:
-        """
-        Loads a tensor by name.
+        """Loads a tensor by name.
 
-        :param name: The name of the tensor to load.
+        Args:
+            name: The name of the tensor to load.
+
+        Returns:
+            The loaded tensor.
         """
         return self._services.tensors.load(name)
 
 
 class ConditioningInterface(InvocationContextInterface):
     def save(self, conditioning_data: ConditioningFieldData) -> str:
-        """
-        Saves a conditioning data object, returning its name.
+        """Saves a conditioning data object, returning its name.
 
-        :param conditioning_data: The conditioning data to save.
+        Args:
+            conditioning_data: The conditioning data to save.
+
+        Returns:
+            The name of the saved conditioning data.
         """
 
         name = self._services.conditioning.save(obj=conditioning_data)
         return name
 
     def load(self, name: str) -> ConditioningFieldData:
-        """
-        Loads conditioning data by name.
+        """Loads conditioning data by name.
 
-        :param name: The name of the conditioning data to load.
+        Args:
+            name: The name of the conditioning data to load.
+
+        Returns:
+            The loaded conditioning data.
         """
 
         return self._services.conditioning.load(name)
 
 
 class ModelsInterface(InvocationContextInterface):
-    def exists(self, key: str) -> bool:
-        """
-        Checks if a model exists.
+    def exists(self, identifier: Union[str, "ModelIdentifierField"]) -> bool:
+        """Checks if a model exists.
 
-        :param key: The key of the model.
-        """
-        return self._services.model_manager.store.exists(key)
+        Args:
+            identifier: The key or ModelField representing the model.
 
-    def load(self, key: str, submodel_type: Optional[SubModelType] = None) -> LoadedModel:
+        Returns:
+            True if the model exists, False if not.
         """
-        Loads a model.
+        if isinstance(identifier, str):
+            return self._services.model_manager.store.exists(identifier)
 
-        :param key: The key of the model.
-        :param submodel_type: The submodel of the model to get.
-        :returns: An object representing the loaded model.
+        return self._services.model_manager.store.exists(identifier.key)
+
+    def load(
+        self, identifier: Union[str, "ModelIdentifierField"], submodel_type: Optional[SubModelType] = None
+    ) -> LoadedModel:
+        """Loads a model.
+
+        Args:
+            identifier: The key or ModelField representing the model.
+            submodel_type: The submodel of the model to get.
+
+        Returns:
+            An object representing the loaded model.
         """
 
         # The model manager emits events as it loads the model. It needs the context data to build
         # the event payloads.
 
-        return self._services.model_manager.load_model_by_key(
-            key=key, submodel_type=submodel_type, context_data=self._data
-        )
+        if isinstance(identifier, str):
+            model = self._services.model_manager.store.get_model(identifier)
+            return self._services.model_manager.load.load_model(model, submodel_type, self._data)
+        else:
+            _submodel_type = submodel_type or identifier.submodel_type
+            model = self._services.model_manager.store.get_model(identifier.key)
+            return self._services.model_manager.load.load_model(model, _submodel_type, self._data)
 
     def load_by_attrs(
-        self, model_name: str, base_model: BaseModelType, model_type: ModelType, submodel: Optional[SubModelType] = None
+        self, name: str, base: BaseModelType, type: ModelType, submodel_type: Optional[SubModelType] = None
     ) -> LoadedModel:
-        """
-        Loads a model by its attributes.
+        """Loads a model by its attributes.
 
-        :param model_name: Name of to be fetched.
-        :param base_model: Base model
-        :param model_type: Type of the model
-        :param submodel: For main (pipeline models), the submodel to fetch
-        """
-        return self._services.model_manager.load_model_by_attr(
-            model_name=model_name,
-            base_model=base_model,
-            model_type=model_type,
-            submodel=submodel,
-            context_data=self._data,
-        )
+        Args:
+            name: Name of the model.
+            base: The models' base type, e.g. `BaseModelType.StableDiffusion1`, `BaseModelType.StableDiffusionXL`, etc.
+            type: Type of the model, e.g. `ModelType.Main`, `ModelType.Vae`, etc.
+            submodel_type: The type of submodel to load, e.g. `SubModelType.UNet`, `SubModelType.TextEncoder`, etc. Only main
+            models have submodels.
 
-    def get_config(self, key: str) -> AnyModelConfig:
+        Returns:
+            An object representing the loaded model.
         """
-        Gets a model's info, an dict-like object.
 
-        :param key: The key of the model.
-        """
-        return self._services.model_manager.store.get_model(key=key)
+        configs = self._services.model_manager.store.search_by_attr(model_name=name, base_model=base, model_type=type)
+        if len(configs) == 0:
+            raise UnknownModelException(f"No model found with name {name}, base {base}, and type {type}")
 
-    def get_metadata(self, key: str) -> Optional[AnyModelRepoMetadata]:
-        """
-        Gets a model's metadata, if it has any.
+        if len(configs) > 1:
+            raise ValueError(f"More than one model found with name {name}, base {base}, and type {type}")
 
-        :param key: The key of the model.
+        return self._services.model_manager.load.load_model(configs[0], submodel_type, self._data)
+
+    def get_config(self, identifier: Union[str, "ModelIdentifierField"]) -> AnyModelConfig:
+        """Gets a model's config.
+
+        Args:
+            identifier: The key or ModelField representing the model.
+
+        Returns:
+            The model's config.
         """
-        return self._services.model_manager.store.get_metadata(key=key)
+        if isinstance(identifier, str):
+            return self._services.model_manager.store.get_model(identifier)
+
+        return self._services.model_manager.store.get_model(identifier.key)
 
     def search_by_path(self, path: Path) -> list[AnyModelConfig]:
-        """
-        Searches for models by path.
+        """Searches for models by path.
 
-        :param path: The path to search for.
+        Args:
+            path: The path to search for.
+
+        Returns:
+            A list of models that match the path.
         """
         return self._services.model_manager.store.search_by_path(path)
 
     def search_by_attrs(
         self,
-        model_name: Optional[str] = None,
-        base_model: Optional[BaseModelType] = None,
-        model_type: Optional[ModelType] = None,
-        model_format: Optional[ModelFormat] = None,
+        name: Optional[str] = None,
+        base: Optional[BaseModelType] = None,
+        type: Optional[ModelType] = None,
+        format: Optional[ModelFormat] = None,
     ) -> list[AnyModelConfig]:
-        """
-        Searches for models by attributes.
+        """Searches for models by attributes.
 
-        :param model_name: Name of to be fetched.
-        :param base_model: Base model
-        :param model_type: Type of the model
-        :param submodel: For main (pipeline models), the submodel to fetch
+        Args:
+            name: The name to search for (exact match).
+            base: The base to search for, e.g. `BaseModelType.StableDiffusion1`, `BaseModelType.StableDiffusionXL`, etc.
+            type: Type type of model to search for, e.g. `ModelType.Main`, `ModelType.Vae`, etc.
+            format: The format of model to search for, e.g. `ModelFormat.Checkpoint`, `ModelFormat.Diffusers`, etc.
+
+        Returns:
+            A list of models that match the attributes.
         """
 
         return self._services.model_manager.store.search_by_attr(
-            model_name=model_name,
-            base_model=base_model,
-            model_type=model_type,
-            model_format=model_format,
+            model_name=name,
+            base_model=base,
+            model_type=type,
+            model_format=format,
         )
 
 
 class ConfigInterface(InvocationContextInterface):
     def get(self) -> InvokeAIAppConfig:
-        """Gets the app's config."""
+        """Gets the app's config.
+
+        Returns:
+            The app's config.
+        """
 
         return self._services.configuration.get_config()
 
@@ -370,7 +434,11 @@ class UtilInterface(InvocationContextInterface):
         self._cancel_event = cancel_event
 
     def is_canceled(self) -> bool:
-        """Checks if the current invocation has been canceled."""
+        """Checks if the current session has been canceled.
+
+        Returns:
+            True if the current session has been canceled, False if not.
+        """
         return self._cancel_event.is_set()
 
     def sd_step_callback(self, intermediate_state: PipelineIntermediateState, base_model: BaseModelType) -> None:
@@ -380,8 +448,9 @@ class UtilInterface(InvocationContextInterface):
 
         This should be called after each denoising step.
 
-        :param intermediate_state: The intermediate state of the diffusion pipeline.
-        :param base_model: The base model for the current denoising step.
+        Args:
+            intermediate_state: The intermediate state of the diffusion pipeline.
+            base_model: The base model for the current denoising step.
         """
 
         stable_diffusion_step_callback(
@@ -394,8 +463,17 @@ class UtilInterface(InvocationContextInterface):
 
 
 class InvocationContext:
-    """
-    The `InvocationContext` provides access to various services and data for the current invocation.
+    """Provides access to various services and data for the current invocation.
+
+    Attributes:
+        images (ImagesInterface): Methods to save, get and update images and their metadata.
+        tensors (TensorsInterface): Methods to save and get tensors, including image, noise, masks, and masked images.
+        conditioning (ConditioningInterface): Methods to save and get conditioning data.
+        models (ModelsInterface): Methods to check if a model exists, get a model, and get a model's info.
+        logger (LoggerInterface): The app logger.
+        config (ConfigInterface): The app config.
+        util (UtilInterface): Utility methods, including a method to check if an invocation was canceled and step callbacks.
+        boards (BoardsInterface): Methods to interact with boards.
     """
 
     def __init__(
@@ -438,11 +516,14 @@ def build_invocation_context(
     data: InvocationContextData,
     cancel_event: threading.Event,
 ) -> InvocationContext:
-    """
-    Builds the invocation context for a specific invocation execution.
+    """Builds the invocation context for a specific invocation execution.
 
-    :param services: The invocation services to wrap.
-    :param data: The invocation context data.
+    Args:
+        services: The invocation services to wrap.
+        data: The invocation context data.
+
+    Returns:
+        The invocation context.
     """
 
     logger = LoggerInterface(services=services, data=data)

invokeai/app/services/shared/sqlite/sqlite_util.py

@@ -9,6 +9,7 @@ from invokeai.app.services.shared.sqlite_migrator.migrations.migration_3 import
 from invokeai.app.services.shared.sqlite_migrator.migrations.migration_4 import build_migration_4
 from invokeai.app.services.shared.sqlite_migrator.migrations.migration_5 import build_migration_5
 from invokeai.app.services.shared.sqlite_migrator.migrations.migration_6 import build_migration_6
+from invokeai.app.services.shared.sqlite_migrator.migrations.migration_7 import build_migration_7
 from invokeai.app.services.shared.sqlite_migrator.sqlite_migrator_impl import SqliteMigrator
 
 
@@ -35,6 +36,7 @@ def init_db(config: InvokeAIAppConfig, logger: Logger, image_files: ImageFileSto
     migrator.register_migration(build_migration_4())
     migrator.register_migration(build_migration_5())
     migrator.register_migration(build_migration_6())
+    migrator.register_migration(build_migration_7())
     migrator.run_migrations()
 
     return db

invokeai/app/services/shared/sqlite_migrator/migrations/migration_3.py

@@ -4,8 +4,6 @@ from logging import Logger
 from invokeai.app.services.config import InvokeAIAppConfig
 from invokeai.app.services.shared.sqlite_migrator.sqlite_migrator_common import Migration
 
-from .util.migrate_yaml_config_1 import MigrateModelYamlToDb1
-
 
 class Migration3Callback:
     def __init__(self, app_config: InvokeAIAppConfig, logger: Logger) -> None:
@@ -15,7 +13,6 @@ class Migration3Callback:
     def __call__(self, cursor: sqlite3.Cursor) -> None:
         self._drop_model_manager_metadata(cursor)
         self._recreate_model_config(cursor)
-        self._migrate_model_config_records(cursor)
 
     def _drop_model_manager_metadata(self, cursor: sqlite3.Cursor) -> None:
         """Drops the `model_manager_metadata` table."""
@@ -55,12 +52,6 @@ class Migration3Callback:
             """
         )
 
-    def _migrate_model_config_records(self, cursor: sqlite3.Cursor) -> None:
-        """After updating the model config table, we repopulate it."""
-        self._logger.info("Migrating model config records from models.yaml to database")
-        model_record_migrator = MigrateModelYamlToDb1(self._app_config, self._logger, cursor)
-        model_record_migrator.migrate()
-
 
 def build_migration_3(app_config: InvokeAIAppConfig, logger: Logger) -> Migration:
     """

invokeai/app/services/shared/sqlite_migrator/migrations/migration_7.py

@@ -0,0 +1,88 @@
+import sqlite3
+
+from invokeai.app.services.shared.sqlite_migrator.sqlite_migrator_common import Migration
+
+
+class Migration7Callback:
+    def __call__(self, cursor: sqlite3.Cursor) -> None:
+        self._create_models_table(cursor)
+        self._drop_old_models_tables(cursor)
+
+    def _drop_old_models_tables(self, cursor: sqlite3.Cursor) -> None:
+        """Drops the old model_records, model_metadata, model_tags and tags tables."""
+
+        tables = ["model_records", "model_metadata", "model_tags", "tags"]
+
+        for table in tables:
+            cursor.execute(f"DROP TABLE IF EXISTS {table};")
+
+    def _create_models_table(self, cursor: sqlite3.Cursor) -> None:
+        """Creates the v4.0.0 models table."""
+
+        tables = [
+            """--sql
+            CREATE TABLE IF NOT EXISTS models (
+                id TEXT NOT NULL PRIMARY KEY,
+                hash TEXT GENERATED ALWAYS as (json_extract(config, '$.hash')) VIRTUAL NOT NULL,
+                base TEXT GENERATED ALWAYS as (json_extract(config, '$.base')) VIRTUAL NOT NULL,
+                type TEXT GENERATED ALWAYS as (json_extract(config, '$.type')) VIRTUAL NOT NULL,
+                path TEXT GENERATED ALWAYS as (json_extract(config, '$.path')) VIRTUAL NOT NULL,
+                format TEXT GENERATED ALWAYS as (json_extract(config, '$.format')) VIRTUAL NOT NULL,
+                name TEXT GENERATED ALWAYS as (json_extract(config, '$.name')) VIRTUAL NOT NULL,
+                description TEXT GENERATED ALWAYS as (json_extract(config, '$.description')) VIRTUAL,
+                source TEXT GENERATED ALWAYS as (json_extract(config, '$.source')) VIRTUAL NOT NULL,
+                source_type TEXT GENERATED ALWAYS as (json_extract(config, '$.source_type')) VIRTUAL NOT NULL,
+                source_api_response TEXT GENERATED ALWAYS as (json_extract(config, '$.source_api_response')) VIRTUAL,
+                trigger_phrases TEXT GENERATED ALWAYS as (json_extract(config, '$.trigger_phrases')) VIRTUAL,
+                -- Serialized JSON representation of the whole config object, which will contain additional fields from subclasses
+                config TEXT NOT NULL,
+                created_at DATETIME NOT NULL DEFAULT(STRFTIME('%Y-%m-%d %H:%M:%f', 'NOW')),
+                -- Updated via trigger
+                updated_at DATETIME NOT NULL DEFAULT(STRFTIME('%Y-%m-%d %H:%M:%f', 'NOW')),
+                -- unique constraint on combo of name, base and type
+                UNIQUE(name, base, type)
+            );
+            """
+        ]
+
+        # Add trigger for `updated_at`.
+        triggers = [
+            """--sql
+            CREATE TRIGGER IF NOT EXISTS models_updated_at
+            AFTER UPDATE
+            ON models FOR EACH ROW
+            BEGIN
+                UPDATE models SET updated_at = STRFTIME('%Y-%m-%d %H:%M:%f', 'NOW')
+                    WHERE id = old.id;
+            END;
+            """
+        ]
+
+        # Add indexes for searchable fields
+        indices = [
+            "CREATE INDEX IF NOT EXISTS base_index ON models(base);",
+            "CREATE INDEX IF NOT EXISTS type_index ON models(type);",
+            "CREATE INDEX IF NOT EXISTS name_index ON models(name);",
+            "CREATE UNIQUE INDEX IF NOT EXISTS path_index ON models(path);",
+        ]
+
+        for stmt in tables + indices + triggers:
+            cursor.execute(stmt)
+
+
+def build_migration_7() -> Migration:
+    """
+    Build the migration from database version 6 to 7.
+
+    This migration does the following:
+    - Adds the new models table
+    - Drops the old model_records, model_metadata, model_tags and tags tables.
+    - TODO(MM2): Migrates model names and descriptions from `models.yaml` to the new table (?).
+    """
+    migration_7 = Migration(
+        from_version=6,
+        to_version=7,
+        callback=Migration7Callback(),
+    )
+
+    return migration_7

invokeai/app/services/shared/sqlite_migrator/migrations/util/migrate_yaml_config_1.py

@@ -1,151 +0,0 @@
-# Copyright (c) 2023 Lincoln D. Stein
-"""Migrate from the InvokeAI v2 models.yaml format to the v3 sqlite format."""
-
-import json
-import sqlite3
-from hashlib import sha1
-from logging import Logger
-from pathlib import Path
-from typing import Optional
-
-from omegaconf import DictConfig, OmegaConf
-from pydantic import TypeAdapter
-
-from invokeai.app.services.config import InvokeAIAppConfig
-from invokeai.app.services.model_records import (
-    DuplicateModelException,
-    UnknownModelException,
-)
-from invokeai.backend.model_manager.config import (
-    AnyModelConfig,
-    BaseModelType,
-    ModelConfigFactory,
-    ModelType,
-)
-from invokeai.backend.model_manager.hash import FastModelHash
-
-ModelsValidator = TypeAdapter(AnyModelConfig)
-
-
-class MigrateModelYamlToDb1:
-    """
-    Migrate the InvokeAI models.yaml format (VERSION 3.0.0) to SQL3 database format (VERSION 3.5.0).
-
-    The class has one externally useful method, migrate(), which scans the
-    currently models.yaml file and imports all its entries into invokeai.db.
-
-    Use this way:
-
-      from invokeai.backend.model_manager/migrate_to_db import MigrateModelYamlToDb
-      MigrateModelYamlToDb().migrate()
-
-    """
-
-    config: InvokeAIAppConfig
-    logger: Logger
-    cursor: sqlite3.Cursor
-
-    def __init__(self, config: InvokeAIAppConfig, logger: Logger, cursor: sqlite3.Cursor = None) -> None:
-        self.config = config
-        self.logger = logger
-        self.cursor = cursor
-
-    def get_yaml(self) -> DictConfig:
-        """Fetch the models.yaml DictConfig for this installation."""
-        yaml_path = self.config.model_conf_path
-        omegaconf = OmegaConf.load(yaml_path)
-        assert isinstance(omegaconf, DictConfig)
-        return omegaconf
-
-    def migrate(self) -> None:
-        """Do the migration from models.yaml to invokeai.db."""
-        try:
-            yaml = self.get_yaml()
-        except OSError:
-            return
-
-        for model_key, stanza in yaml.items():
-            if model_key == "__metadata__":
-                assert (
-                    stanza["version"] == "3.0.0"
-                ), f"This script works on version 3.0.0 yaml files, but your configuration points to a {stanza['version']} version"
-                continue
-
-            base_type, model_type, model_name = str(model_key).split("/")
-            try:
-                hash = FastModelHash.hash(self.config.models_path / stanza.path)
-            except OSError:
-                self.logger.warning(f"The model at {stanza.path} is not a valid file or directory. Skipping migration.")
-                continue
-
-            assert isinstance(model_key, str)
-            new_key = sha1(model_key.encode("utf-8")).hexdigest()
-
-            stanza["base"] = BaseModelType(base_type)
-            stanza["type"] = ModelType(model_type)
-            stanza["name"] = model_name
-            stanza["original_hash"] = hash
-            stanza["current_hash"] = hash
-
-            new_config: AnyModelConfig = ModelsValidator.validate_python(stanza)  # type: ignore # see https://github.com/pydantic/pydantic/discussions/7094
-
-            try:
-                if original_record := self._search_by_path(stanza.path):
-                    key = original_record.key
-                    self.logger.info(f"Updating model {model_name} with information from models.yaml using key {key}")
-                    self._update_model(key, new_config)
-                else:
-                    self.logger.info(f"Adding model {model_name} with key {model_key}")
-                    self._add_model(new_key, new_config)
-            except DuplicateModelException:
-                self.logger.warning(f"Model {model_name} is already in the database")
-            except UnknownModelException:
-                self.logger.warning(f"Model at {stanza.path} could not be found in database")
-
-    def _search_by_path(self, path: Path) -> Optional[AnyModelConfig]:
-        self.cursor.execute(
-            """--sql
-            SELECT config FROM model_config
-            WHERE path=?;
-            """,
-            (str(path),),
-        )
-        results = [ModelConfigFactory.make_config(json.loads(x[0])) for x in self.cursor.fetchall()]
-        return results[0] if results else None
-
-    def _update_model(self, key: str, config: AnyModelConfig) -> None:
-        record = ModelConfigFactory.make_config(config, key=key)  # ensure it is a valid config obect
-        json_serialized = record.model_dump_json()  # and turn it into a json string.
-        self.cursor.execute(
-            """--sql
-            UPDATE model_config
-            SET
-                config=?
-            WHERE id=?;
-            """,
-            (json_serialized, key),
-        )
-        if self.cursor.rowcount == 0:
-            raise UnknownModelException("model not found")
-
-    def _add_model(self, key: str, config: AnyModelConfig) -> None:
-        record = ModelConfigFactory.make_config(config, key=key)  # ensure it is a valid config obect.
-        json_serialized = record.model_dump_json()  # and turn it into a json string.
-        try:
-            self.cursor.execute(
-                """--sql
-                INSERT INTO model_config (
-                   id,
-                   original_hash,
-                   config
-                  )
-                VALUES (?,?,?);
-                """,
-                (
-                    key,
-                    record.original_hash,
-                    json_serialized,
-                ),
-            )
-        except sqlite3.IntegrityError as exc:
-            raise DuplicateModelException(f"{record.name}: model is already in database") from exc

invokeai/app/services/shared/sqlite_migrator/sqlite_migrator_common.py

@@ -17,8 +17,7 @@ class MigrateCallback(Protocol):
     See :class:`Migration` for an example.
     """
 
-    def __call__(self, cursor: sqlite3.Cursor) -> None:
-        ...
+    def __call__(self, cursor: sqlite3.Cursor) -> None: ...
 
 
 class MigrationError(RuntimeError):

invokeai/app/services/urls/urls_base.py

@@ -8,3 +8,8 @@ class UrlServiceBase(ABC):
     def get_image_url(self, image_name: str, thumbnail: bool = False) -> str:
         """Gets the URL for an image or thumbnail."""
         pass
+
+    @abstractmethod
+    def get_model_image_url(self, model_key: str) -> str:
+        """Gets the URL for a model image"""
+        pass

invokeai/app/services/urls/urls_default.py

@@ -4,8 +4,9 @@ from .urls_base import UrlServiceBase
 
 
 class LocalUrlService(UrlServiceBase):
-    def __init__(self, base_url: str = "api/v1"):
+    def __init__(self, base_url: str = "api/v1", base_url_v2: str = "api/v2"):
         self._base_url = base_url
+        self._base_url_v2 = base_url_v2
 
     def get_image_url(self, image_name: str, thumbnail: bool = False) -> str:
         image_basename = os.path.basename(image_name)
@@ -15,3 +16,6 @@ class LocalUrlService(UrlServiceBase):
             return f"{self._base_url}/images/i/{image_basename}/thumbnail"
 
         return f"{self._base_url}/images/i/{image_basename}/full"
+
+    def get_model_image_url(self, model_key: str) -> str:
+        return f"{self._base_url_v2}/models/i/{model_key}/image"

invokeai/app/util/metadata.py

@@ -1,55 +0,0 @@
-import json
-from typing import Optional
-
-from pydantic import ValidationError
-
-from invokeai.app.services.shared.graph import Edge
-
-
-def get_metadata_graph_from_raw_session(session_raw: str) -> Optional[dict]:
-    """
-    Parses raw session string, returning a dict of the graph.
-
-    Only the general graph shape is validated; none of the fields are validated.
-
-    Any `metadata_accumulator` nodes and edges are removed.
-
-    Any validation failure will return None.
-    """
-
-    graph = json.loads(session_raw).get("graph", None)
-
-    # sanity check make sure the graph is at least reasonably shaped
-    if (
-        not isinstance(graph, dict)
-        or "nodes" not in graph
-        or not isinstance(graph["nodes"], dict)
-        or "edges" not in graph
-        or not isinstance(graph["edges"], list)
-    ):
-        # something has gone terribly awry, return an empty dict
-        return None
-
-    try:
-        # delete the `metadata_accumulator` node
-        del graph["nodes"]["metadata_accumulator"]
-    except KeyError:
-        # no accumulator node, all good
-        pass
-
-    # delete any edges to or from it
-    for i, edge in enumerate(graph["edges"]):
-        try:
-            # try to parse the edge
-            Edge(**edge)
-        except ValidationError:
-            # something has gone terribly awry, return an empty dict
-            return None
-
-        if (
-            edge["source"]["node_id"] == "metadata_accumulator"
-            or edge["destination"]["node_id"] == "metadata_accumulator"
-        ):
-            del graph["edges"][i]
-
-    return graph

invokeai/app/util/ti_utils.py

@@ -22,7 +22,7 @@ def generate_ti_list(
     for trigger in extract_ti_triggers_from_prompt(prompt):
         name_or_key = trigger[1:-1]
         try:
-            loaded_model = context.models.load(key=name_or_key)
+            loaded_model = context.models.load(name_or_key)
             model = loaded_model.model
             assert isinstance(model, TextualInversionModelRaw)
             assert loaded_model.config.base == base
@@ -30,7 +30,7 @@ def generate_ti_list(
         except UnknownModelException:
             try:
                 loaded_model = context.models.load_by_attrs(
-                    model_name=name_or_key, base_model=base, model_type=ModelType.TextualInversion
+                    name=name_or_key, base=base, type=ModelType.TextualInversion
                 )
                 model = loaded_model.model
                 assert isinstance(model, TextualInversionModelRaw)

invokeai/backend/image_util/__init__.py

@@ -1,6 +1,7 @@
 """
 Initialization file for invokeai.backend.image_util methods.
 """
+
 from .patchmatch import PatchMatch  # noqa: F401
 from .pngwriter import PngWriter, PromptFormatter, retrieve_metadata, write_metadata  # noqa: F401
 from .seamless import configure_model_padding  # noqa: F401

invokeai/backend/image_util/depth_anything/__init__.py

@@ -13,9 +13,11 @@ from invokeai.app.services.config.config_default import InvokeAIAppConfig
 from invokeai.backend.image_util.depth_anything.model.dpt import DPT_DINOv2
 from invokeai.backend.image_util.depth_anything.utilities.util import NormalizeImage, PrepareForNet, Resize
 from invokeai.backend.util.devices import choose_torch_device
+from invokeai.backend.util.logging import InvokeAILogger
 from invokeai.backend.util.util import download_with_progress_bar
 
 config = InvokeAIAppConfig.get_config()
+logger = InvokeAILogger.get_logger(config=config)
 
 DEPTH_ANYTHING_MODELS = {
     "large": {
@@ -54,8 +56,9 @@ class DepthAnythingDetector:
     def __init__(self) -> None:
         self.model = None
         self.model_size: Union[Literal["large", "base", "small"], None] = None
+        self.device = choose_torch_device()
 
-    def load_model(self, model_size=Literal["large", "base", "small"]):
+    def load_model(self, model_size: Literal["large", "base", "small"] = "small"):
         DEPTH_ANYTHING_MODEL_PATH = pathlib.Path(config.models_path / DEPTH_ANYTHING_MODELS[model_size]["local"])
         if not DEPTH_ANYTHING_MODEL_PATH.exists():
             download_with_progress_bar(DEPTH_ANYTHING_MODELS[model_size]["url"], DEPTH_ANYTHING_MODEL_PATH)
@@ -71,8 +74,6 @@ class DepthAnythingDetector:
                     self.model = DPT_DINOv2(encoder="vitb", features=128, out_channels=[96, 192, 384, 768])
                 case "large":
                     self.model = DPT_DINOv2(encoder="vitl", features=256, out_channels=[256, 512, 1024, 1024])
-                case _:
-                    raise TypeError("Not a supported model")
 
             self.model.load_state_dict(torch.load(DEPTH_ANYTHING_MODEL_PATH.as_posix(), map_location="cpu"))
             self.model.eval()
@@ -80,20 +81,20 @@ class DepthAnythingDetector:
         self.model.to(choose_torch_device())
         return self.model
 
-    def to(self, device):
-        self.model.to(device)
-        return self
+    def __call__(self, image: Image.Image, resolution: int = 512) -> Image.Image:
+        if not self.model:
+            logger.warn("DepthAnything model was not loaded. Returning original image")
+            return image
 
-    def __call__(self, image, resolution=512, offload=False):
-        image = np.array(image, dtype=np.uint8)
-        image = image[:, :, ::-1] / 255.0
+        np_image = np.array(image, dtype=np.uint8)
+        np_image = np_image[:, :, ::-1] / 255.0
 
-        image_height, image_width = image.shape[:2]
-        image = transform({"image": image})["image"]
-        image = torch.from_numpy(image).unsqueeze(0).to(choose_torch_device())
+        image_height, image_width = np_image.shape[:2]
+        np_image = transform({"image": np_image})["image"]
+        tensor_image = torch.from_numpy(np_image).unsqueeze(0).to(choose_torch_device())
 
         with torch.no_grad():
-            depth = self.model(image)
+            depth = self.model(tensor_image)
             depth = F.interpolate(depth[None], (image_height, image_width), mode="bilinear", align_corners=False)[0, 0]
             depth = (depth - depth.min()) / (depth.max() - depth.min()) * 255.0
 
@@ -103,7 +104,4 @@ class DepthAnythingDetector:
         new_height = int(image_height * (resolution / image_width))
         depth_map = depth_map.resize((resolution, new_height))
 
-        if offload:
-            del self.model
-
         return depth_map

invokeai/backend/image_util/invisible_watermark.py

@@ -3,6 +3,7 @@ This module defines a singleton object, "invisible_watermark" that
 wraps the invisible watermark model. It respects the global "invisible_watermark"
 configuration variable, that allows the watermarking to be supressed.
 """
+
 import cv2
 import numpy as np
 from imwatermark import WatermarkEncoder

invokeai/backend/image_util/patchmatch.py

@@ -4,6 +4,7 @@ wraps the actual patchmatch object. It respects the global
 "try_patchmatch" attribute, so that patchmatch loading can
 be suppressed or deferred
 """
+
 import numpy as np
 
 import invokeai.backend.util.logging as logger

invokeai/backend/image_util/pngwriter.py

@@ -6,6 +6,7 @@ PngWriter -- Converts Images generated by T2I into PNGs, finds
 
 Exports function retrieve_metadata(path)
 """
+
 import json
 import os
 import re

invokeai/backend/image_util/safety_checker.py

@@ -3,6 +3,7 @@ This module defines a singleton object, "safety_checker" that
 wraps the safety_checker model. It respects the global "nsfw_checker"
 configuration variable, that allows the checker to be supressed.
 """
+
 import numpy as np
 from PIL import Image
 

invokeai/backend/install/check_root.py

@@ -1,6 +1,7 @@
 """
 Check that the invokeai_root is correctly configured and exit if not.
 """
+
 import sys
 
 from invokeai.app.services.config import InvokeAIAppConfig
@@ -10,17 +11,6 @@ def check_invokeai_root(config: InvokeAIAppConfig):
     try:
         assert config.db_path.parent.exists(), f"{config.db_path.parent} not found"
         assert config.models_path.exists(), f"{config.models_path} not found"
-        if not config.ignore_missing_core_models:
-            for model in [
-                "CLIP-ViT-bigG-14-laion2B-39B-b160k",
-                "bert-base-uncased",
-                "clip-vit-large-patch14",
-                "sd-vae-ft-mse",
-                "stable-diffusion-2-clip",
-                "stable-diffusion-safety-checker",
-            ]:
-                path = config.models_path / f"core/convert/{model}"
-                assert path.exists(), f"{path} is missing"
     except Exception as e:
         print()
         print(f"An exception has occurred: {str(e)}")
@@ -31,10 +21,5 @@ def check_invokeai_root(config: InvokeAIAppConfig):
         print(
             '** From the command line, activate the virtual environment and run "invokeai-configure --yes --skip-sd-weights" **'
         )
-        print(
-            '** (To skip this check completely, add "--ignore_missing_core_models" to your CLI args. Not installing '
-            "these core models will prevent the loading of some or all .safetensors and .ckpt files. However, you can "
-            "always come back and install these core models in the future.)"
-        )
         input("Press any key to continue...")
         sys.exit(0)

invokeai/backend/install/install_helper.py

@@ -1,4 +1,5 @@
 """Utility (backend) functions used by model_install.py"""
+
 from logging import Logger
 from pathlib import Path
 from typing import Any, Dict, List, Optional
@@ -18,7 +19,6 @@ from invokeai.app.services.model_install import (
     ModelInstallService,
     ModelInstallServiceBase,
 )
-from invokeai.app.services.model_metadata import ModelMetadataStoreSQL
 from invokeai.app.services.model_records import ModelRecordServiceBase, ModelRecordServiceSQL
 from invokeai.app.services.shared.sqlite.sqlite_util import init_db
 from invokeai.backend.model_manager import (
@@ -38,7 +38,7 @@ def initialize_record_store(app_config: InvokeAIAppConfig) -> ModelRecordService
     logger = InvokeAILogger.get_logger(config=app_config)
     image_files = DiskImageFileStorage(f"{app_config.output_path}/images")
     db = init_db(config=app_config, logger=logger, image_files=image_files)
-    obj: ModelRecordServiceBase = ModelRecordServiceSQL(db, ModelMetadataStoreSQL(db))
+    obj: ModelRecordServiceBase = ModelRecordServiceSQL(db)
     return obj
 
 

invokeai/backend/install/invokeai_configure.py

@@ -17,7 +17,7 @@ import warnings
 from argparse import Namespace
 from enum import Enum
 from pathlib import Path
-from shutil import get_terminal_size
+from shutil import copy, get_terminal_size, move
 from typing import Any, Optional, Set, Tuple, Type, get_args, get_type_hints
 from urllib import request
 
@@ -25,20 +25,20 @@ import npyscreen
 import psutil
 import torch
 import transformers
-from diffusers import AutoencoderKL, ModelMixin
+from diffusers import ModelMixin
 from diffusers.pipelines.stable_diffusion.safety_checker import StableDiffusionSafetyChecker
 from huggingface_hub import HfFolder
 from huggingface_hub import login as hf_hub_login
 from omegaconf import DictConfig, OmegaConf
 from pydantic.error_wrappers import ValidationError
 from tqdm import tqdm
-from transformers import AutoFeatureExtractor, BertTokenizerFast, CLIPTextConfig, CLIPTextModel, CLIPTokenizer
+from transformers import AutoFeatureExtractor
 
 import invokeai.configs as configs
 from invokeai.app.services.config import InvokeAIAppConfig
 from invokeai.backend.install.install_helper import InstallHelper, InstallSelections
 from invokeai.backend.install.legacy_arg_parsing import legacy_parser
-from invokeai.backend.model_manager import BaseModelType, ModelType
+from invokeai.backend.model_manager import ModelType
 from invokeai.backend.util import choose_precision, choose_torch_device
 from invokeai.backend.util.logging import InvokeAILogger
 from invokeai.frontend.install.model_install import addModelsForm
@@ -210,51 +210,15 @@ def download_with_progress_bar(model_url: str, model_dest: str, label: str = "th
         print(traceback.format_exc(), file=sys.stderr)
 
 
-def download_conversion_models():
+def download_safety_checker():
     target_dir = config.models_path / "core/convert"
     kwargs = {}  # for future use
     try:
-        logger.info("Downloading core tokenizers and text encoders")
-
-        # bert
-        with warnings.catch_warnings():
-            warnings.filterwarnings("ignore", category=DeprecationWarning)
-            bert = BertTokenizerFast.from_pretrained("bert-base-uncased", **kwargs)
-            bert.save_pretrained(target_dir / "bert-base-uncased", safe_serialization=True)
-
-        # sd-1
-        repo_id = "openai/clip-vit-large-patch14"
-        hf_download_from_pretrained(CLIPTokenizer, repo_id, target_dir / "clip-vit-large-patch14")
-        hf_download_from_pretrained(CLIPTextModel, repo_id, target_dir / "clip-vit-large-patch14")
-
-        # sd-2
-        repo_id = "stabilityai/stable-diffusion-2"
-        pipeline = CLIPTokenizer.from_pretrained(repo_id, subfolder="tokenizer", **kwargs)
-        pipeline.save_pretrained(target_dir / "stable-diffusion-2-clip" / "tokenizer", safe_serialization=True)
-
-        pipeline = CLIPTextModel.from_pretrained(repo_id, subfolder="text_encoder", **kwargs)
-        pipeline.save_pretrained(target_dir / "stable-diffusion-2-clip" / "text_encoder", safe_serialization=True)
-
-        # sd-xl - tokenizer_2
-        repo_id = "laion/CLIP-ViT-bigG-14-laion2B-39B-b160k"
-        _, model_name = repo_id.split("/")
-        pipeline = CLIPTokenizer.from_pretrained(repo_id, **kwargs)
-        pipeline.save_pretrained(target_dir / model_name, safe_serialization=True)
-
-        pipeline = CLIPTextConfig.from_pretrained(repo_id, **kwargs)
-        pipeline.save_pretrained(target_dir / model_name, safe_serialization=True)
-
-        # VAE
-        logger.info("Downloading stable diffusion VAE")
-        vae = AutoencoderKL.from_pretrained("stabilityai/sd-vae-ft-mse", **kwargs)
-        vae.save_pretrained(target_dir / "sd-vae-ft-mse", safe_serialization=True)
-
         # safety checking
         logger.info("Downloading safety checker")
         repo_id = "CompVis/stable-diffusion-safety-checker"
         pipeline = AutoFeatureExtractor.from_pretrained(repo_id, **kwargs)
         pipeline.save_pretrained(target_dir / "stable-diffusion-safety-checker", safe_serialization=True)
-
         pipeline = StableDiffusionSafetyChecker.from_pretrained(repo_id, **kwargs)
         pipeline.save_pretrained(target_dir / "stable-diffusion-safety-checker", safe_serialization=True)
     except KeyboardInterrupt:
@@ -307,7 +271,7 @@ def download_lama():
 def download_support_models() -> None:
     download_realesrgan()
     download_lama()
-    download_conversion_models()
+    download_safety_checker()
 
 
 # -------------------------------------
@@ -744,12 +708,7 @@ def initialize_rootdir(root: Path, yes_to_all: bool = False):
         shutil.copytree(configs_src, configs_dest, dirs_exist_ok=True)
 
     dest = root / "models"
-    for model_base in BaseModelType:
-        for model_type in ModelType:
-            path = dest / model_base.value / model_type.value
-            path.mkdir(parents=True, exist_ok=True)
-    path = dest / "core"
-    path.mkdir(parents=True, exist_ok=True)
+    dest.mkdir(parents=True, exist_ok=True)
 
 
 # -------------------------------------
@@ -929,6 +888,10 @@ def main() -> None:
 
     errors = set()
     FORCE_FULL_PRECISION = opt.full_precision  # FIXME global
+    new_init_file = config.root_path / "invokeai.yaml"
+    backup_init_file = new_init_file.with_suffix(".bak")
+    if new_init_file.exists():
+        copy(new_init_file, backup_init_file)
 
     try:
         # if we do a root migration/upgrade, then we are keeping previous
@@ -943,7 +906,6 @@ def main() -> None:
         install_helper = InstallHelper(config, logger)
 
         models_to_download = default_user_selections(opt, install_helper)
-        new_init_file = config.root_path / "invokeai.yaml"
 
         if opt.yes_to_all:
             write_default_options(opt, new_init_file)
@@ -975,8 +937,17 @@ def main() -> None:
             input("Press any key to continue...")
     except WindowTooSmallException as e:
         logger.error(str(e))
+        if backup_init_file.exists():
+            move(backup_init_file, new_init_file)
     except KeyboardInterrupt:
         print("\nGoodbye! Come back soon.")
+        if backup_init_file.exists():
+            move(backup_init_file, new_init_file)
+    except Exception:
+        print("An error occurred during installation.")
+        if backup_init_file.exists():
+            move(backup_init_file, new_init_file)
+        print(traceback.format_exc(), file=sys.stderr)
 
 
 # -------------------------------------

invokeai/backend/model_hash/model_hash.py

@@ -0,0 +1,219 @@
+# Copyright (c) 2023 Lincoln D. Stein and the InvokeAI Development Team
+
+import hashlib
+import os
+from pathlib import Path
+from typing import Callable, Literal, Optional, Union
+
+from blake3 import blake3
+
+from invokeai.app.util.misc import uuid_string
+
+HASHING_ALGORITHMS = Literal[
+    "md5",
+    "sha1",
+    "sha224",
+    "sha256",
+    "sha384",
+    "sha512",
+    "blake2b",
+    "blake2s",
+    "sha3_224",
+    "sha3_256",
+    "sha3_384",
+    "sha3_512",
+    "shake_128",
+    "shake_256",
+    "blake3",
+    "blake3_single",
+    "random",
+]
+MODEL_FILE_EXTENSIONS = (".ckpt", ".safetensors", ".bin", ".pt", ".pth")
+
+
+class ModelHash:
+    """
+    Creates a hash of a model using a specified algorithm. The hash is prefixed by the algorithm used.
+
+    Args:
+        algorithm: Hashing algorithm to use. Defaults to BLAKE3.
+        file_filter: A function that takes a file name and returns True if the file should be included in the hash.
+
+    If the model is a single file, it is hashed directly using the provided algorithm.
+
+    If the model is a directory, each model weights file in the directory is hashed using the provided algorithm.
+
+    Only files with the following extensions are hashed: .ckpt, .safetensors, .bin, .pt, .pth
+
+    The final hash is computed by hashing the hashes of all model files in the directory using BLAKE3, ensuring
+    that directory hashes are never weaker than the file hashes.
+
+    A convenience algorithm choice of "random" is also available, which returns a random string. This is not a hash.
+
+    Usage:
+        ```py
+        # BLAKE3 hash
+        ModelHash().hash("path/to/some/model.safetensors") # "blake3:ce3f0c5f3c05d119f4a5dcaf209b50d3149046a0d3a9adee9fed4c83cad6b4d0"
+        # MD5
+        ModelHash("md5").hash("path/to/model/dir/") # "md5:a0cd925fc063f98dbf029eee315060c3"
+        ```
+    """
+
+    def __init__(
+        self, algorithm: HASHING_ALGORITHMS = "blake3", file_filter: Optional[Callable[[str], bool]] = None
+    ) -> None:
+        self.algorithm: HASHING_ALGORITHMS = algorithm
+        if algorithm == "blake3":
+            self._hash_file = self._blake3
+        elif algorithm == "blake3_single":
+            self._hash_file = self._blake3_single
+        elif algorithm in hashlib.algorithms_available:
+            self._hash_file = self._get_hashlib(algorithm)
+        elif algorithm == "random":
+            self._hash_file = self._random
+        else:
+            raise ValueError(f"Algorithm {algorithm} not available")
+
+        self._file_filter = file_filter or self._default_file_filter
+
+    def hash(self, model_path: Union[str, Path]) -> str:
+        """
+        Return hexdigest of hash of model located at model_path using the algorithm provided at class instantiation.
+
+        If model_path is a directory, the hash is computed by hashing the hashes of all model files in the
+        directory. The final composite hash is always computed using BLAKE3.
+
+        Args:
+            model_path: Path to the model
+
+        Returns:
+            str: Hexdigest of the hash of the model
+        """
+
+        model_path = Path(model_path)
+        # blake3_single is a single-threaded version of blake3, prefix should still be "blake3:"
+        prefix = self._get_prefix(self.algorithm)
+        if model_path.is_file():
+            return prefix + self._hash_file(model_path)
+        elif model_path.is_dir():
+            return prefix + self._hash_dir(model_path)
+        else:
+            raise OSError(f"Not a valid file or directory: {model_path}")
+
+    def _hash_dir(self, dir: Path) -> str:
+        """Compute the hash for all files in a directory and return a hexdigest.
+
+        Args:
+            dir: Path to the directory
+
+        Returns:
+            str: Hexdigest of the hash of the directory
+        """
+        model_component_paths = self._get_file_paths(dir, self._file_filter)
+
+        component_hashes: list[str] = []
+        for component in sorted(model_component_paths):
+            component_hashes.append(self._hash_file(component))
+
+        # BLAKE3 is cryptographically secure. We may as well fall back on a secure algorithm
+        # for the composite hash
+        composite_hasher = blake3()
+        for h in component_hashes:
+            composite_hasher.update(h.encode("utf-8"))
+
+        return composite_hasher.hexdigest()
+
+    @staticmethod
+    def _get_file_paths(model_path: Path, file_filter: Callable[[str], bool]) -> list[Path]:
+        """Return a list of all model files in the directory.
+
+        Args:
+            model_path: Path to the model
+            file_filter: Function that takes a file name and returns True if the file should be included in the list.
+
+        Returns:
+            List of all model files in the directory
+        """
+
+        files: list[Path] = []
+        for root, _dirs, _files in os.walk(model_path):
+            for file in _files:
+                if file_filter(file):
+                    files.append(Path(root, file))
+        return files
+
+    @staticmethod
+    def _blake3(file_path: Path) -> str:
+        """Hashes a file using BLAKE3, using parallelized and memory-mapped I/O to avoid reading the entire file into memory.
+
+        Args:
+            file_path: Path to the file to hash
+
+        Returns:
+            Hexdigest of the hash of the file
+        """
+        file_hasher = blake3(max_threads=blake3.AUTO)
+        file_hasher.update_mmap(file_path)
+        return file_hasher.hexdigest()
+
+    @staticmethod
+    def _blake3_single(file_path: Path) -> str:
+        """Hashes a file using BLAKE3, without parallelism. Suitable for spinning hard drives.
+
+        Args:
+            file_path: Path to the file to hash
+
+        Returns:
+            Hexdigest of the hash of the file
+        """
+        file_hasher = blake3()
+        file_hasher.update_mmap(file_path)
+        return file_hasher.hexdigest()
+
+    @staticmethod
+    def _get_hashlib(algorithm: HASHING_ALGORITHMS) -> Callable[[Path], str]:
+        """Factory function that returns a function to hash a file with the given algorithm.
+
+        Args:
+            algorithm: Hashing algorithm to use
+
+        Returns:
+            A function that hashes a file using the given algorithm
+        """
+
+        def hashlib_hasher(file_path: Path) -> str:
+            """Hashes a file using a hashlib algorithm. Uses `memoryview` to avoid reading the entire file into memory."""
+            hasher = hashlib.new(algorithm)
+            buffer = bytearray(128 * 1024)
+            mv = memoryview(buffer)
+            with open(file_path, "rb", buffering=0) as f:
+                while n := f.readinto(mv):
+                    hasher.update(mv[:n])
+            return hasher.hexdigest()
+
+        return hashlib_hasher
+
+    @staticmethod
+    def _random(_file_path: Path) -> str:
+        """Returns a random string. This is not a hash.
+
+        The string is a UUID, hashed with BLAKE3 to ensure that it is unique."""
+        return blake3(uuid_string().encode()).hexdigest()
+
+    @staticmethod
+    def _default_file_filter(file_path: str) -> bool:
+        """A default file filter that only includes files with the following extensions: .ckpt, .safetensors, .bin, .pt, .pth
+
+        Args:
+            file_path: Path to the file
+
+        Returns:
+            True if the file matches the given extensions, otherwise False
+        """
+        return file_path.endswith(MODEL_FILE_EXTENSIONS)
+
+    @staticmethod
+    def _get_prefix(algorithm: HASHING_ALGORITHMS) -> str:
+        """Return the prefix for the given algorithm, e.g. \"blake3:\" or \"md5:\"."""
+        # blake3_single is a single-threaded version of blake3, prefix should still be "blake3:"
+        return "blake3:" if algorithm == "blake3_single" else f"{algorithm}:"

invokeai/backend/model_manager/__init__.py

@@ -1,4 +1,5 @@
 """Re-export frequently-used symbols from the Model Manager backend."""
+
 from .config import (
     AnyModel,
     AnyModelConfig,

invokeai/backend/model_manager/config.py

@@ -19,15 +19,19 @@ Typical usage:
 Validation errors will raise an InvalidModelConfigException error.
 
 """
+
 import time
 from enum import Enum
-from typing import Literal, Optional, Type, Union
+from typing import Literal, Optional, Type, TypeAlias, Union
 
 import torch
-from diffusers import ModelMixin
-from pydantic import BaseModel, ConfigDict, Field, TypeAdapter
+from diffusers.models.modeling_utils import ModelMixin
+from pydantic import BaseModel, ConfigDict, Discriminator, Field, Tag, TypeAdapter
 from typing_extensions import Annotated, Any, Dict
 
+from invokeai.app.invocations.constants import SCHEDULER_NAME_VALUES
+from invokeai.app.util.misc import uuid_string
+
 from ..raw_model import RawModel
 
 # ModelMixin is the base class for all diffusers and transformers models
@@ -55,8 +59,8 @@ class ModelType(str, Enum):
 
     ONNX = "onnx"
     Main = "main"
-    Vae = "vae"
-    Lora = "lora"
+    VAE = "vae"
+    LoRA = "lora"
     ControlNet = "controlnet"  # used by model_probe
     TextualInversion = "embedding"
     IPAdapter = "ip_adapter"
@@ -72,9 +76,9 @@ class SubModelType(str, Enum):
     TextEncoder2 = "text_encoder_2"
     Tokenizer = "tokenizer"
     Tokenizer2 = "tokenizer_2"
-    Vae = "vae"
-    VaeDecoder = "vae_decoder"
-    VaeEncoder = "vae_encoder"
+    VAE = "vae"
+    VAEDecoder = "vae_decoder"
+    VAEEncoder = "vae_encoder"
     Scheduler = "scheduler"
     SafetyChecker = "safety_checker"
 
@@ -92,8 +96,8 @@ class ModelFormat(str, Enum):
 
     Diffusers = "diffusers"
     Checkpoint = "checkpoint"
-    Lycoris = "lycoris"
-    Onnx = "onnx"
+    LyCORIS = "lycoris"
+    ONNX = "onnx"
     Olive = "olive"
     EmbeddingFile = "embedding_file"
     EmbeddingFolder = "embedding_folder"
@@ -111,127 +115,208 @@ class SchedulerPredictionType(str, Enum):
 class ModelRepoVariant(str, Enum):
     """Various hugging face variants on the diffusers format."""
 
-    DEFAULT = ""  # model files without "fp16" or other qualifier - empty str
+    Default = ""  # model files without "fp16" or other qualifier - empty str
     FP16 = "fp16"
     FP32 = "fp32"
     ONNX = "onnx"
-    OPENVINO = "openvino"
-    FLAX = "flax"
+    OpenVINO = "openvino"
+    Flax = "flax"
+
+
+class ModelSourceType(str, Enum):
+    """Model source type."""
+
+    Path = "path"
+    Url = "url"
+    HFRepoID = "hf_repo_id"
+
+
+DEFAULTS_PRECISION = Literal["fp16", "fp32"]
+
+
+class MainModelDefaultSettings(BaseModel):
+    vae: str | None = Field(default=None, description="Default VAE for this model (model key)")
+    vae_precision: DEFAULTS_PRECISION | None = Field(default=None, description="Default VAE precision for this model")
+    scheduler: SCHEDULER_NAME_VALUES | None = Field(default=None, description="Default scheduler for this model")
+    steps: int | None = Field(default=None, gt=0, description="Default number of steps for this model")
+    cfg_scale: float | None = Field(default=None, ge=1, description="Default CFG Scale for this model")
+    cfg_rescale_multiplier: float | None = Field(
+        default=None, ge=0, lt=1, description="Default CFG Rescale Multiplier for this model"
+    )
+    width: int | None = Field(default=None, multiple_of=8, ge=64, description="Default width for this model")
+    height: int | None = Field(default=None, multiple_of=8, ge=64, description="Default height for this model")
+
+
+class ControlAdapterDefaultSettings(BaseModel):
+    # This could be narrowed to controlnet processor nodes, but they change. Leaving this a string is safer.
+    preprocessor: str | None
 
 
 class ModelConfigBase(BaseModel):
     """Base class for model configuration information."""
 
-    path: str = Field(description="filesystem path to the model file or directory")
-    name: str = Field(description="model name")
-    base: BaseModelType = Field(description="base model")
-    type: ModelType = Field(description="type of the model")
-    format: ModelFormat = Field(description="model format")
-    key: str = Field(description="unique key for model", default="<NOKEY>")
-    original_hash: Optional[str] = Field(
-        description="original fasthash of model contents", default=None
-    )  # this is assigned at install time and will not change
-    current_hash: Optional[str] = Field(
-        description="current fasthash of model contents", default=None
-    )  # if model is converted or otherwise modified, this will hold updated hash
-    description: Optional[str] = Field(description="human readable description of the model", default=None)
-    source: Optional[str] = Field(description="model original source (path, URL or repo_id)", default=None)
-    last_modified: Optional[float] = Field(description="timestamp for modification time", default_factory=time.time)
+    key: str = Field(description="A unique key for this model.", default_factory=uuid_string)
+    hash: str = Field(description="The hash of the model file(s).")
+    path: str = Field(
+        description="Path to the model on the filesystem. Relative paths are relative to the Invoke root directory."
+    )
+    name: str = Field(description="Name of the model.")
+    base: BaseModelType = Field(description="The base model.")
+    description: Optional[str] = Field(description="Model description", default=None)
+    source: str = Field(description="The original source of the model (path, URL or repo_id).")
+    source_type: ModelSourceType = Field(description="The type of source")
+    source_api_response: Optional[str] = Field(
+        description="The original API response from the source, as stringified JSON.", default=None
+    )
+    cover_image: Optional[str] = Field(description="Url for image to preview model", default=None)
 
     @staticmethod
     def json_schema_extra(schema: dict[str, Any], model_class: Type[BaseModel]) -> None:
-        schema["required"].extend(
-            ["key", "base", "type", "format", "original_hash", "current_hash", "source", "last_modified"]
-        )
+        schema["required"].extend(["key", "type", "format"])
 
-    model_config = ConfigDict(
-        use_enum_values=False,
-        validate_assignment=True,
-        json_schema_extra=json_schema_extra,
-    )
-
-    def update(self, attributes: Dict[str, Any]) -> None:
-        """Update the object with fields in dict."""
-        for key, value in attributes.items():
-            setattr(self, key, value)  # may raise a validation error
+    model_config = ConfigDict(validate_assignment=True, json_schema_extra=json_schema_extra)
 
 
-class _CheckpointConfig(ModelConfigBase):
+class CheckpointConfigBase(ModelConfigBase):
     """Model config for checkpoint-style models."""
 
     format: Literal[ModelFormat.Checkpoint] = ModelFormat.Checkpoint
-    config: str = Field(description="path to the checkpoint model config file")
+    config_path: str = Field(description="path to the checkpoint model config file")
+    converted_at: Optional[float] = Field(
+        description="When this model was last converted to diffusers", default_factory=time.time
+    )
 
 
-class _DiffusersConfig(ModelConfigBase):
+class DiffusersConfigBase(ModelConfigBase):
     """Model config for diffusers-style models."""
 
     format: Literal[ModelFormat.Diffusers] = ModelFormat.Diffusers
-    repo_variant: Optional[ModelRepoVariant] = ModelRepoVariant.DEFAULT
+    repo_variant: Optional[ModelRepoVariant] = ModelRepoVariant.Default
 
 
-class LoRAConfig(ModelConfigBase):
+class LoRAConfigBase(ModelConfigBase):
+    type: Literal[ModelType.LoRA] = ModelType.LoRA
+    trigger_phrases: Optional[set[str]] = Field(description="Set of trigger phrases for this model", default=None)
+
+
+class LoRALyCORISConfig(LoRAConfigBase):
     """Model config for LoRA/Lycoris models."""
 
-    type: Literal[ModelType.Lora] = ModelType.Lora
-    format: Literal[ModelFormat.Lycoris, ModelFormat.Diffusers]
+    format: Literal[ModelFormat.LyCORIS] = ModelFormat.LyCORIS
+
+    @staticmethod
+    def get_tag() -> Tag:
+        return Tag(f"{ModelType.LoRA.value}.{ModelFormat.LyCORIS.value}")
 
 
-class VaeCheckpointConfig(ModelConfigBase):
+class LoRADiffusersConfig(LoRAConfigBase):
+    """Model config for LoRA/Diffusers models."""
+
+    format: Literal[ModelFormat.Diffusers] = ModelFormat.Diffusers
+
+    @staticmethod
+    def get_tag() -> Tag:
+        return Tag(f"{ModelType.LoRA.value}.{ModelFormat.Diffusers.value}")
+
+
+class VAECheckpointConfig(CheckpointConfigBase):
     """Model config for standalone VAE models."""
 
-    type: Literal[ModelType.Vae] = ModelType.Vae
+    type: Literal[ModelType.VAE] = ModelType.VAE
     format: Literal[ModelFormat.Checkpoint] = ModelFormat.Checkpoint
 
+    @staticmethod
+    def get_tag() -> Tag:
+        return Tag(f"{ModelType.VAE.value}.{ModelFormat.Checkpoint.value}")
 
-class VaeDiffusersConfig(ModelConfigBase):
+
+class VAEDiffusersConfig(ModelConfigBase):
     """Model config for standalone VAE models (diffusers version)."""
 
-    type: Literal[ModelType.Vae] = ModelType.Vae
+    type: Literal[ModelType.VAE] = ModelType.VAE
     format: Literal[ModelFormat.Diffusers] = ModelFormat.Diffusers
 
+    @staticmethod
+    def get_tag() -> Tag:
+        return Tag(f"{ModelType.VAE.value}.{ModelFormat.Diffusers.value}")
 
-class ControlNetDiffusersConfig(_DiffusersConfig):
+
+class ControlAdapterConfigBase(BaseModel):
+    default_settings: Optional[ControlAdapterDefaultSettings] = Field(
+        description="Default settings for this model", default=None
+    )
+
+
+class ControlNetDiffusersConfig(DiffusersConfigBase, ControlAdapterConfigBase):
     """Model config for ControlNet models (diffusers version)."""
 
     type: Literal[ModelType.ControlNet] = ModelType.ControlNet
     format: Literal[ModelFormat.Diffusers] = ModelFormat.Diffusers
 
+    @staticmethod
+    def get_tag() -> Tag:
+        return Tag(f"{ModelType.ControlNet.value}.{ModelFormat.Diffusers.value}")
 
-class ControlNetCheckpointConfig(_CheckpointConfig):
+
+class ControlNetCheckpointConfig(CheckpointConfigBase, ControlAdapterConfigBase):
     """Model config for ControlNet models (diffusers version)."""
 
     type: Literal[ModelType.ControlNet] = ModelType.ControlNet
     format: Literal[ModelFormat.Checkpoint] = ModelFormat.Checkpoint
 
+    @staticmethod
+    def get_tag() -> Tag:
+        return Tag(f"{ModelType.ControlNet.value}.{ModelFormat.Checkpoint.value}")
 
-class TextualInversionConfig(ModelConfigBase):
+
+class TextualInversionFileConfig(ModelConfigBase):
     """Model config for textual inversion embeddings."""
 
     type: Literal[ModelType.TextualInversion] = ModelType.TextualInversion
-    format: Literal[ModelFormat.EmbeddingFile, ModelFormat.EmbeddingFolder]
+    format: Literal[ModelFormat.EmbeddingFile] = ModelFormat.EmbeddingFile
+
+    @staticmethod
+    def get_tag() -> Tag:
+        return Tag(f"{ModelType.TextualInversion.value}.{ModelFormat.EmbeddingFile.value}")
 
 
-class _MainConfig(ModelConfigBase):
-    """Model config for main models."""
+class TextualInversionFolderConfig(ModelConfigBase):
+    """Model config for textual inversion embeddings."""
+
+    type: Literal[ModelType.TextualInversion] = ModelType.TextualInversion
+    format: Literal[ModelFormat.EmbeddingFolder] = ModelFormat.EmbeddingFolder
+
+    @staticmethod
+    def get_tag() -> Tag:
+        return Tag(f"{ModelType.TextualInversion.value}.{ModelFormat.EmbeddingFolder.value}")
+
+
+class MainConfigBase(ModelConfigBase):
+    type: Literal[ModelType.Main] = ModelType.Main
+    trigger_phrases: Optional[set[str]] = Field(description="Set of trigger phrases for this model", default=None)
+    default_settings: Optional[MainModelDefaultSettings] = Field(
+        description="Default settings for this model", default=None
+    )
+
+
+class MainCheckpointConfig(CheckpointConfigBase, MainConfigBase):
+    """Model config for main checkpoint models."""
 
-    vae: Optional[str] = Field(default=None)
     variant: ModelVariantType = ModelVariantType.Normal
     prediction_type: SchedulerPredictionType = SchedulerPredictionType.Epsilon
     upcast_attention: bool = False
-    ztsnr_training: bool = False
+
+    @staticmethod
+    def get_tag() -> Tag:
+        return Tag(f"{ModelType.Main.value}.{ModelFormat.Checkpoint.value}")
 
 
-class MainCheckpointConfig(_CheckpointConfig, _MainConfig):
-    """Model config for main checkpoint models."""
-
-    type: Literal[ModelType.Main] = ModelType.Main
-
-
-class MainDiffusersConfig(_DiffusersConfig, _MainConfig):
+class MainDiffusersConfig(DiffusersConfigBase, MainConfigBase):
     """Model config for main diffusers models."""
 
-    type: Literal[ModelType.Main] = ModelType.Main
+    @staticmethod
+    def get_tag() -> Tag:
+        return Tag(f"{ModelType.Main.value}.{ModelFormat.Diffusers.value}")
 
 
 class IPAdapterConfig(ModelConfigBase):
@@ -241,63 +326,75 @@ class IPAdapterConfig(ModelConfigBase):
     image_encoder_model_id: str
     format: Literal[ModelFormat.InvokeAI]
 
+    @staticmethod
+    def get_tag() -> Tag:
+        return Tag(f"{ModelType.IPAdapter.value}.{ModelFormat.InvokeAI.value}")
+
 
 class CLIPVisionDiffusersConfig(ModelConfigBase):
-    """Model config for ClipVision."""
+    """Model config for CLIPVision."""
 
     type: Literal[ModelType.CLIPVision] = ModelType.CLIPVision
     format: Literal[ModelFormat.Diffusers]
 
+    @staticmethod
+    def get_tag() -> Tag:
+        return Tag(f"{ModelType.CLIPVision.value}.{ModelFormat.Diffusers.value}")
 
-class T2IConfig(ModelConfigBase):
+
+class T2IAdapterConfig(ModelConfigBase, ControlAdapterConfigBase):
     """Model config for T2I."""
 
     type: Literal[ModelType.T2IAdapter] = ModelType.T2IAdapter
     format: Literal[ModelFormat.Diffusers]
 
+    @staticmethod
+    def get_tag() -> Tag:
+        return Tag(f"{ModelType.T2IAdapter.value}.{ModelFormat.Diffusers.value}")
 
-_ControlNetConfig = Annotated[
-    Union[ControlNetDiffusersConfig, ControlNetCheckpointConfig],
-    Field(discriminator="format"),
-]
-_VaeConfig = Annotated[Union[VaeDiffusersConfig, VaeCheckpointConfig], Field(discriminator="format")]
-_MainModelConfig = Annotated[Union[MainDiffusersConfig, MainCheckpointConfig], Field(discriminator="format")]
 
-AnyModelConfig = Union[
-    _MainModelConfig,
-    _VaeConfig,
-    _ControlNetConfig,
-    # ModelConfigBase,
-    LoRAConfig,
-    TextualInversionConfig,
-    IPAdapterConfig,
-    CLIPVisionDiffusersConfig,
-    T2IConfig,
+def get_model_discriminator_value(v: Any) -> str:
+    """
+    Computes the discriminator value for a model config.
+    https://docs.pydantic.dev/latest/concepts/unions/#discriminated-unions-with-callable-discriminator
+    """
+    format_ = None
+    type_ = None
+    if isinstance(v, dict):
+        format_ = v.get("format")
+        if isinstance(format_, Enum):
+            format_ = format_.value
+        type_ = v.get("type")
+        if isinstance(type_, Enum):
+            type_ = type_.value
+    else:
+        format_ = v.format.value
+        type_ = v.type.value
+    v = f"{type_}.{format_}"
+    return v
+
+
+AnyModelConfig = Annotated[
+    Union[
+        Annotated[MainDiffusersConfig, MainDiffusersConfig.get_tag()],
+        Annotated[MainCheckpointConfig, MainCheckpointConfig.get_tag()],
+        Annotated[VAEDiffusersConfig, VAEDiffusersConfig.get_tag()],
+        Annotated[VAECheckpointConfig, VAECheckpointConfig.get_tag()],
+        Annotated[ControlNetDiffusersConfig, ControlNetDiffusersConfig.get_tag()],
+        Annotated[ControlNetCheckpointConfig, ControlNetCheckpointConfig.get_tag()],
+        Annotated[LoRALyCORISConfig, LoRALyCORISConfig.get_tag()],
+        Annotated[LoRADiffusersConfig, LoRADiffusersConfig.get_tag()],
+        Annotated[TextualInversionFileConfig, TextualInversionFileConfig.get_tag()],
+        Annotated[TextualInversionFolderConfig, TextualInversionFolderConfig.get_tag()],
+        Annotated[IPAdapterConfig, IPAdapterConfig.get_tag()],
+        Annotated[T2IAdapterConfig, T2IAdapterConfig.get_tag()],
+        Annotated[CLIPVisionDiffusersConfig, CLIPVisionDiffusersConfig.get_tag()],
+    ],
+    Discriminator(get_model_discriminator_value),
 ]
 
 AnyModelConfigValidator = TypeAdapter(AnyModelConfig)
-
-
-# IMPLEMENTATION NOTE:
-# The preferred alternative to the above is a discriminated Union as shown
-# below. However, it breaks FastAPI when used as the input Body parameter in a route.
-# This is a known issue. Please see:
-#   https://github.com/tiangolo/fastapi/discussions/9761 and
-#   https://github.com/tiangolo/fastapi/discussions/9287
-# AnyModelConfig = Annotated[
-#     Union[
-#         _MainModelConfig,
-#         _ONNXConfig,
-#         _VaeConfig,
-#         _ControlNetConfig,
-#         LoRAConfig,
-#         TextualInversionConfig,
-#         IPAdapterConfig,
-#         CLIPVisionDiffusersConfig,
-#         T2IConfig,
-#     ],
-#     Field(discriminator="type"),
-# ]
+AnyDefaultSettings: TypeAlias = Union[MainModelDefaultSettings, ControlAdapterDefaultSettings]
 
 
 class ModelConfigFactory(object):
@@ -331,6 +428,6 @@ class ModelConfigFactory(object):
         assert model is not None
         if key:
             model.key = key
-        if timestamp:
-            model.last_modified = timestamp
+        if isinstance(model, CheckpointConfigBase) and timestamp is not None:
+            model.converted_at = timestamp
         return model  # type: ignore

invokeai/backend/model_manager/hash.py

@@ -1,66 +0,0 @@
-# Copyright (c) 2023 Lincoln D. Stein and the InvokeAI Development Team
-"""
-Fast hashing of diffusers and checkpoint-style models.
-
-Usage:
-from invokeai.backend.model_managre.model_hash import FastModelHash
->>> FastModelHash.hash('/home/models/stable-diffusion-v1.5')
-'a8e693a126ea5b831c96064dc569956f'
-"""
-
-import hashlib
-import os
-from pathlib import Path
-from typing import Dict, Union
-
-from imohash import hashfile
-
-
-class FastModelHash(object):
-    """FastModelHash obect provides one public class method, hash()."""
-
-    @classmethod
-    def hash(cls, model_location: Union[str, Path]) -> str:
-        """
-        Return hexdigest string for model located at model_location.
-
-        :param model_location: Path to the model
-        """
-        model_location = Path(model_location)
-        if model_location.is_file():
-            return cls._hash_file(model_location)
-        elif model_location.is_dir():
-            return cls._hash_dir(model_location)
-        else:
-            raise OSError(f"Not a valid file or directory: {model_location}")
-
-    @classmethod
-    def _hash_file(cls, model_location: Union[str, Path]) -> str:
-        """
-        Fasthash a single file and return its hexdigest.
-
-        :param model_location: Path to the model file
-        """
-        # we return md5 hash of the filehash to make it shorter
-        # cryptographic security not needed here
-        return hashlib.md5(hashfile(model_location)).hexdigest()
-
-    @classmethod
-    def _hash_dir(cls, model_location: Union[str, Path]) -> str:
-        components: Dict[str, str] = {}
-
-        for root, _dirs, files in os.walk(model_location):
-            for file in files:
-                # only tally tensor files because diffusers config files change slightly
-                # depending on how the model was downloaded/converted.
-                if not file.endswith((".ckpt", ".safetensors", ".bin", ".pt", ".pth")):
-                    continue
-                path = (Path(root) / file).as_posix()
-                fast_hash = cls._hash_file(path)
-                components.update({path: fast_hash})
-
-        # hash all the model hashes together, using alphabetic file order
-        md5 = hashlib.md5()
-        for _path, fast_hash in sorted(components.items()):
-            md5.update(fast_hash.encode("utf-8"))
-        return md5.hexdigest()

invokeai/backend/model_manager/load/__init__.py

@@ -2,6 +2,7 @@
 """
 Init file for the model loader.
 """
+
 from importlib import import_module
 from pathlib import Path
 

invokeai/backend/model_manager/load/convert_cache/convert_cache_base.py

@@ -1,6 +1,7 @@
 """
 Disk-based converted model cache.
 """
+
 from abc import ABC, abstractmethod
 from pathlib import Path
 

invokeai/backend/model_manager/load/load_default.py

@@ -13,6 +13,7 @@ from invokeai.backend.model_manager import (
     ModelRepoVariant,
     SubModelType,
 )
+from invokeai.backend.model_manager.config import DiffusersConfigBase, ModelType
 from invokeai.backend.model_manager.load.convert_cache import ModelConvertCacheBase
 from invokeai.backend.model_manager.load.load_base import LoadedModel, ModelLoaderBase
 from invokeai.backend.model_manager.load.model_cache.model_cache_base import ModelCacheBase, ModelLockerBase
@@ -50,7 +51,7 @@ class ModelLoader(ModelLoaderBase):
         :param submodel_type: an ModelType enum indicating the portion of
                the model to retrieve (e.g. ModelType.Vae)
         """
-        if model_config.type == "main" and not submodel_type:
+        if model_config.type is ModelType.Main and not submodel_type:
             raise InvalidModelConfigException("submodel_type is required when loading a main model")
 
         model_path, model_config, submodel_type = self._get_model_path(model_config, submodel_type)
@@ -80,7 +81,7 @@ class ModelLoader(ModelLoaderBase):
         self._convert_cache.make_room(self.get_size_fs(config, model_path, submodel_type))
         return self._convert_model(config, model_path, cache_path)
 
-    def _needs_conversion(self, config: AnyModelConfig, model_path: Path, cache_path: Path) -> bool:
+    def _needs_conversion(self, config: AnyModelConfig, model_path: Path, dest_path: Path) -> bool:
         return False
 
     def _load_if_needed(
@@ -119,7 +120,7 @@ class ModelLoader(ModelLoaderBase):
         return calc_model_size_by_fs(
             model_path=model_path,
             subfolder=submodel_type.value if submodel_type else None,
-            variant=config.repo_variant if hasattr(config, "repo_variant") else None,
+            variant=config.repo_variant if isinstance(config, DiffusersConfigBase) else None,
         )
 
     # This needs to be implemented in subclasses that handle checkpoints

invokeai/backend/model_manager/load/model_cache/model_cache_default.py

@@ -19,7 +19,6 @@ context. Use like this:
 """
 
 import gc
-import logging
 import math
 import sys
 import time
@@ -92,8 +91,7 @@ class ModelCache(ModelCacheBase[AnyModel]):
         self._execution_device: torch.device = execution_device
         self._storage_device: torch.device = storage_device
         self._logger = logger or InvokeAILogger.get_logger(self.__class__.__name__)
-        self._log_memory_usage = log_memory_usage or self._logger.level == logging.DEBUG
-        # used for stats collection
+        self._log_memory_usage = log_memory_usage
         self._stats: Optional[CacheStats] = None
 
         self._cached_models: Dict[str, CacheRecord[AnyModel]] = {}

invokeai/backend/model_manager/load/model_loader_registry.py

@@ -14,10 +14,9 @@ Use like this:
     ).load_model(model_config, submodel_type)
 
 """
-import hashlib
+
 from abc import ABC, abstractmethod
-from pathlib import Path
-from typing import Callable, Dict, Optional, Tuple, Type
+from typing import Callable, Dict, Optional, Tuple, Type, TypeVar
 
 from ..config import (
     AnyModelConfig,
@@ -26,8 +25,6 @@ from ..config import (
     ModelFormat,
     ModelType,
     SubModelType,
-    VaeCheckpointConfig,
-    VaeDiffusersConfig,
 )
 from . import ModelLoaderBase
 
@@ -60,7 +57,10 @@ class ModelLoaderRegistryBase(ABC):
         """
 
 
-class ModelLoaderRegistry:
+TModelLoader = TypeVar("TModelLoader", bound=ModelLoaderBase)
+
+
+class ModelLoaderRegistry(ModelLoaderRegistryBase):
     """
     This class allows model loaders to register their type, base and format.
     """
@@ -70,10 +70,10 @@ class ModelLoaderRegistry:
     @classmethod
     def register(
         cls, type: ModelType, format: ModelFormat, base: BaseModelType = BaseModelType.Any
-    ) -> Callable[[Type[ModelLoaderBase]], Type[ModelLoaderBase]]:
+    ) -> Callable[[Type[TModelLoader]], Type[TModelLoader]]:
         """Define a decorator which registers the subclass of loader."""
 
-        def decorator(subclass: Type[ModelLoaderBase]) -> Type[ModelLoaderBase]:
+        def decorator(subclass: Type[TModelLoader]) -> Type[TModelLoader]:
             key = cls._to_registry_key(base, type, format)
             if key in cls._registry:
                 raise Exception(
@@ -89,33 +89,15 @@ class ModelLoaderRegistry:
         cls, config: AnyModelConfig, submodel_type: Optional[SubModelType]
     ) -> Tuple[Type[ModelLoaderBase], ModelConfigBase, Optional[SubModelType]]:
         """Get subclass of ModelLoaderBase registered to handle base and type."""
-        # We have to handle VAE overrides here because this will change the model type and the corresponding implementation returned
-        conf2, submodel_type = cls._handle_subtype_overrides(config, submodel_type)
 
-        key1 = cls._to_registry_key(conf2.base, conf2.type, conf2.format)  # for a specific base type
-        key2 = cls._to_registry_key(BaseModelType.Any, conf2.type, conf2.format)  # with wildcard Any
+        key1 = cls._to_registry_key(config.base, config.type, config.format)  # for a specific base type
+        key2 = cls._to_registry_key(BaseModelType.Any, config.type, config.format)  # with wildcard Any
         implementation = cls._registry.get(key1) or cls._registry.get(key2)
         if not implementation:
             raise NotImplementedError(
                 f"No subclass of LoadedModel is registered for base={config.base}, type={config.type}, format={config.format}"
             )
-        return implementation, conf2, submodel_type
-
-    @classmethod
-    def _handle_subtype_overrides(
-        cls, config: AnyModelConfig, submodel_type: Optional[SubModelType]
-    ) -> Tuple[ModelConfigBase, Optional[SubModelType]]:
-        if submodel_type == SubModelType.Vae and hasattr(config, "vae") and config.vae is not None:
-            model_path = Path(config.vae)
-            config_class = (
-                VaeCheckpointConfig if model_path.suffix in [".pt", ".safetensors", ".ckpt"] else VaeDiffusersConfig
-            )
-            hash = hashlib.md5(model_path.as_posix().encode("utf-8")).hexdigest()
-            new_conf = config_class(path=model_path.as_posix(), name=model_path.stem, base=config.base, key=hash)
-            submodel_type = None
-        else:
-            new_conf = config
-        return new_conf, submodel_type
+        return implementation, config, submodel_type
 
     @staticmethod
     def _to_registry_key(base: BaseModelType, type: ModelType, format: ModelFormat) -> str:

invokeai/backend/model_manager/load/model_loaders/controlnet.py

@@ -3,15 +3,13 @@
 
 from pathlib import Path
 
-import safetensors
-import torch
-
 from invokeai.backend.model_manager import (
     AnyModelConfig,
     BaseModelType,
     ModelFormat,
     ModelType,
 )
+from invokeai.backend.model_manager.config import CheckpointConfigBase
 from invokeai.backend.model_manager.convert_ckpt_to_diffusers import convert_controlnet_to_diffusers
 
 from .. import ModelLoaderRegistry
@@ -20,15 +18,15 @@ from .generic_diffusers import GenericDiffusersLoader
 
 @ModelLoaderRegistry.register(base=BaseModelType.Any, type=ModelType.ControlNet, format=ModelFormat.Diffusers)
 @ModelLoaderRegistry.register(base=BaseModelType.Any, type=ModelType.ControlNet, format=ModelFormat.Checkpoint)
-class ControlnetLoader(GenericDiffusersLoader):
+class ControlNetLoader(GenericDiffusersLoader):
     """Class to load ControlNet models."""
 
     def _needs_conversion(self, config: AnyModelConfig, model_path: Path, dest_path: Path) -> bool:
-        if config.format != ModelFormat.Checkpoint:
+        if not isinstance(config, CheckpointConfigBase):
             return False
         elif (
             dest_path.exists()
-            and (dest_path / "config.json").stat().st_mtime >= (config.last_modified or 0.0)
+            and (dest_path / "config.json").stat().st_mtime >= (config.converted_at or 0.0)
             and (dest_path / "config.json").stat().st_mtime >= model_path.stat().st_mtime
         ):
             return False
@@ -36,27 +34,25 @@ class ControlnetLoader(GenericDiffusersLoader):
             return True
 
     def _convert_model(self, config: AnyModelConfig, model_path: Path, output_path: Path) -> Path:
-        if config.base not in {BaseModelType.StableDiffusion1, BaseModelType.StableDiffusion2}:
-            raise Exception(f"Vae conversion not supported for model type: {config.base}")
-        else:
-            assert hasattr(config, "config")
-            config_file = config.config
+        assert isinstance(config, CheckpointConfigBase)
+        config_file = config.config_path
 
-        if model_path.suffix == ".safetensors":
-            checkpoint = safetensors.torch.load_file(model_path, device="cpu")
-        else:
-            checkpoint = torch.load(model_path, map_location="cpu")
-
-        # sometimes weights are hidden under "state_dict", and sometimes not
-        if "state_dict" in checkpoint:
-            checkpoint = checkpoint["state_dict"]
+        image_size = (
+            512
+            if config.base == BaseModelType.StableDiffusion1
+            else 768
+            if config.base == BaseModelType.StableDiffusion2
+            else 1024
+        )
 
+        self._logger.info(f"Converting {model_path} to diffusers format")
+        with open(self._app_config.root_path / config_file, "r") as config_stream:
             convert_controlnet_to_diffusers(
                 model_path,
                 output_path,
-            original_config_file=self._app_config.root_path / config_file,
-            image_size=512,
-            scan_needed=True,
+                original_config_file=config_stream,
+                image_size=image_size,
+                precision=self._torch_dtype,
                 from_safetensors=model_path.suffix == ".safetensors",
             )
         return output_path

invokeai/backend/model_manager/load/model_loaders/generic_diffusers.py

@@ -3,9 +3,10 @@
 
 import sys
 from pathlib import Path
-from typing import Any, Dict, Optional
+from typing import Any, Optional
 
-from diffusers import ConfigMixin, ModelMixin
+from diffusers.configuration_utils import ConfigMixin
+from diffusers.models.modeling_utils import ModelMixin
 
 from invokeai.backend.model_manager import (
     AnyModel,
@@ -41,6 +42,7 @@ class GenericDiffusersLoader(ModelLoader):
     # TO DO: Add exception handling
     def get_hf_load_class(self, model_path: Path, submodel_type: Optional[SubModelType] = None) -> ModelMixin:
         """Given the model path and submodel, returns the diffusers ModelMixin subclass needed to load."""
+        result = None
         if submodel_type:
             try:
                 config = self._load_diffusers_config(model_path, config_name="model_index.json")
@@ -64,6 +66,7 @@ class GenericDiffusersLoader(ModelLoader):
                     raise InvalidModelConfigException("Unable to decifer Load Class based on given config.json")
             except KeyError as e:
                 raise InvalidModelConfigException("An expected config.json file is missing from this model.") from e
+        assert result is not None
         return result
 
     # TO DO: Add exception handling
@@ -75,7 +78,7 @@ class GenericDiffusersLoader(ModelLoader):
         result: ModelMixin = getattr(res_type, class_name)
         return result
 
-    def _load_diffusers_config(self, model_path: Path, config_name: str = "config.json") -> Dict[str, Any]:
+    def _load_diffusers_config(self, model_path: Path, config_name: str = "config.json") -> dict[str, Any]:
         return ConfigLoader.load_config(model_path, config_name=config_name)
 
 
@@ -83,8 +86,8 @@ class ConfigLoader(ConfigMixin):
     """Subclass of ConfigMixin for loading diffusers configuration files."""
 
     @classmethod
-    def load_config(cls, *args: Any, **kwargs: Any) -> Dict[str, Any]:
+    def load_config(cls, *args: Any, **kwargs: Any) -> dict[str, Any]:  # pyright: ignore [reportIncompatibleMethodOverride]
         """Load a diffusrs ConfigMixin configuration."""
         cls.config_name = kwargs.pop("config_name")
-        # Diffusers doesn't provide typing info
+        # TODO(psyche): the types on this diffusers method are not correct
         return super().load_config(*args, **kwargs)  # type: ignore