final tidying before marking PR as ready for review

- Replace AnyModelLoader with ModelLoaderRegistry
- Fix type check errors in multiple files
- Remove apparently unneeded `get_model_config_enum()` method from model manager
- Remove last vestiges of old model manager
- Updated tests and documentation

resolve conflict with seamless.py

.github/CODEOWNERS

@@ -1,5 +1,5 @@
 # continuous integration
-/.github/workflows/  @lstein @blessedcoolant @hipsterusername
+/.github/workflows/  @lstein @blessedcoolant @hipsterusername @ebr
 
 # documentation
 /docs/ @lstein @blessedcoolant @hipsterusername @Millu
@@ -10,7 +10,7 @@
 
 # installation and configuration
 /pyproject.toml  @lstein @blessedcoolant @hipsterusername
-/docker/  @lstein @blessedcoolant @hipsterusername
+/docker/  @lstein @blessedcoolant @hipsterusername @ebr
 /scripts/ @ebr @lstein @hipsterusername
 /installer/ @lstein @ebr @hipsterusername
 /invokeai/assets @lstein @ebr @hipsterusername
@@ -26,9 +26,7 @@
 
 # front ends
 /invokeai/frontend/CLI @lstein @hipsterusername
-/invokeai/frontend/install @lstein @ebr @hipsterusername 
+/invokeai/frontend/install @lstein @ebr @hipsterusername
 /invokeai/frontend/merge @lstein @blessedcoolant @hipsterusername
 /invokeai/frontend/training @lstein @blessedcoolant @hipsterusername
 /invokeai/frontend/web @psychedelicious @blessedcoolant @maryhipp @hipsterusername
-
-

.github/ISSUE_TEMPLATE/BUG_REPORT.yml

@@ -6,10 +6,6 @@ title: '[bug]: '
 
 labels: ['bug']
 
-# assignees:
-#   - moderator_bot
-#   - lstein
-
 body:
   - type: markdown
     attributes:
@@ -18,10 +14,9 @@ body:
 
   - type: checkboxes
     attributes:
-      label: Is there an existing issue for this?
+      label: Is there an existing issue for this problem?
       description: |
-        Please use the [search function](https://github.com/invoke-ai/InvokeAI/issues?q=is%3Aissue+is%3Aopen+label%3Abug)
-        irst to see if an issue already exists for the bug you encountered.
+        Please [search](https://github.com/invoke-ai/InvokeAI/issues) first to see if an issue already exists for the problem.
       options:
         - label: I have searched the existing issues
           required: true
@@ -33,80 +28,119 @@ body:
   - type: dropdown
     id: os_dropdown
     attributes:
-      label: OS
-      description: Which operating System did you use when the bug occured
+      label: Operating system
+      description: Your computer's operating system.
       multiple: false
       options:
         - 'Linux'
         - 'Windows'
         - 'macOS'
+        - 'other'
     validations:
       required: true
 
   - type: dropdown
     id: gpu_dropdown
     attributes:
-      label: GPU
-      description: Which kind of Graphic-Adapter is your System using
+      label: GPU vendor
+      description: Your GPU's vendor.
       multiple: false
       options:
-        - 'cuda'
-        - 'amd'
-        - 'mps'
-        - 'cpu'
+        - 'Nvidia (CUDA)'
+        - 'AMD (ROCm)'
+        - 'Apple Silicon (MPS)'
+        - 'None (CPU)'
     validations:
       required: true
 
+  - type: input
+    id: gpu_model
+    attributes:
+      label: GPU model
+      description: Your GPU's model. If on Apple Silicon, this is your Mac's chip. Leave blank if on CPU.
+      placeholder: ex. RTX 2080 Ti, Mac M1 Pro
+    validations:
+      required: false
+
   - type: input
     id: vram
     attributes:
-      label: VRAM
-      description: Size of the VRAM if known
+      label: GPU VRAM
+      description: Your GPU's VRAM. If on Apple Silicon, this is your Mac's unified memory. Leave blank if on CPU.
       placeholder: 8GB
     validations:
       required: false
-      
+
   - type: input
     id: version-number
     attributes:
-      label: What version did you experience this issue on?
+      label: Version number
       description: |
-        Please share the version of Invoke AI that you experienced the issue on. If this is not the latest version, please update first to confirm the issue still exists. If you are testing main, please include the commit hash instead.
-      placeholder: X.X.X
+        The version of Invoke you have installed. If it is not the latest version, please update and try again to confirm the issue still exists. If you are testing main, please include the commit hash instead.
+      placeholder: ex. 3.6.1
     validations:
       required: true
 
+  - type: input
+    id: browser-version
+    attributes:
+      label: Browser
+      description: Your web browser and version.
+      placeholder: ex. Firefox 123.0b3
+    validations:
+      required: true
+
+  - type: textarea
+    id: python-deps
+    attributes:
+      label: Python dependencies
+      description: |
+        If the problem occurred during image generation, click the gear icon at the bottom left corner, click "About", click the copy button and then paste here.
+    validations:
+      required: false
+
   - type: textarea
     id: what-happened
     attributes:
-      label: What happened?
+      label: What happened
       description: |
-        Briefly describe what happened, what you expected to happen and how to reproduce this bug.
-      placeholder: When using the webinterface and right-clicking on button X instead of the popup-menu there error Y appears
+        Describe what happened. Include any relevant error messages, stack traces and screenshots here.
+      placeholder: I clicked button X and then Y happened.
     validations:
       required: true
 
   - type: textarea
+    id: what-you-expected
     attributes:
-      label: Screenshots
-      description: If applicable, add screenshots to help explain your problem
-      placeholder: this is what the result looked like <screenshot>
+      label: What you expected to happen
+      description: Describe what you expected to happen.
+      placeholder: I expected Z to happen.
+    validations:
+      required: true
+
+  - type: textarea
+    id: how-to-repro
+    attributes:
+      label: How to reproduce the problem
+      description: List steps to reproduce the problem.
+      placeholder: Start the app, generate an image with these settings, then click button X.
     validations:
       required: false
 
   - type: textarea
+    id: additional-context
     attributes:
       label: Additional context
-      description: Add any other context about the problem here
+      description: Any other context that might help us to understand the problem.
       placeholder: Only happens when there is full moon and Friday the 13th on Christmas Eve 🎅🏻
     validations:
       required: false
 
   - type: input
-    id: contact
+    id: discord-username
     attributes:
-      label: Contact Details
-      description: __OPTIONAL__ How can we get in touch with you if we need more info (besides this issue)?
-      placeholder: ex. email@example.com, discordname, twitter, ...
+      label: Discord username
+      description: If you are on the Invoke discord and would prefer to be contacted there, please provide your username.
+      placeholder: supercoolusername123
     validations:
       required: false

.github/pr_labels.yml

@@ -0,0 +1,59 @@
+Root:
+- changed-files:
+  - any-glob-to-any-file: '*'
+
+PythonDeps:
+- changed-files:
+  - any-glob-to-any-file: 'pyproject.toml'
+
+Python:
+- changed-files:
+  - all-globs-to-any-file: 
+    - 'invokeai/**'
+    - '!invokeai/frontend/web/**'
+
+PythonTests:
+- changed-files:
+  - any-glob-to-any-file: 'tests/**'
+
+CICD:
+- changed-files:
+  - any-glob-to-any-file: .github/**
+
+Docker:
+- changed-files:
+  - any-glob-to-any-file: docker/**
+
+Installer:
+- changed-files:
+  - any-glob-to-any-file: installer/**
+
+Documentation:
+- changed-files:
+  - any-glob-to-any-file: docs/**
+
+Invocations:
+- changed-files:
+  - any-glob-to-any-file: 'invokeai/app/invocations/**'
+
+Backend:
+- changed-files:
+  - any-glob-to-any-file: 'invokeai/backend/**'
+
+Api:
+- changed-files:
+  - any-glob-to-any-file: 'invokeai/app/api/**'
+
+Services:
+- changed-files:
+  - any-glob-to-any-file: 'invokeai/app/services/**'
+
+FrontendDeps:
+- changed-files:
+  - any-glob-to-any-file:
+    - '**/*/package.json'
+    - '**/*/pnpm-lock.yaml'
+
+Frontend:
+- changed-files:
+  - any-glob-to-any-file: 'invokeai/frontend/web/**'

.github/workflows/build-container.yml

@@ -40,10 +40,14 @@ jobs:
       - name: Free up more disk space on the runner
         # https://github.com/actions/runner-images/issues/2840#issuecomment-1284059930
         run: |
+          echo "----- Free space before cleanup"
+          df -h
           sudo rm -rf /usr/share/dotnet
           sudo rm -rf "$AGENT_TOOLSDIRECTORY"
           sudo swapoff /mnt/swapfile
           sudo rm -rf /mnt/swapfile
+          echo "----- Free space after cleanup"
+          df -h
 
       - name: Checkout
         uses: actions/checkout@v3
@@ -91,6 +95,7 @@ jobs:
       #     password: ${{ secrets.DOCKERHUB_TOKEN }}
 
       - name: Build container
+        timeout-minutes: 40
         id: docker_build
         uses: docker/build-push-action@v4
         with:

.github/workflows/label-pr.yml

@@ -0,0 +1,16 @@
+name: "Pull Request Labeler"
+on:
+- pull_request_target
+
+jobs:
+  labeler:
+    permissions:
+      contents: read
+      pull-requests: write
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+      - uses: actions/labeler@v5
+        with:
+          configuration-path: .github/pr_labels.yml

.github/workflows/lint-frontend.yml

@@ -21,16 +21,16 @@ jobs:
     if: github.event.pull_request.draft == false
     runs-on: ubuntu-22.04
     steps:
-      - name: Setup Node 20
+      - name: Setup Node 18
         uses: actions/setup-node@v4
         with:
-          node-version: '20'
+          node-version: '18'
       - name: Checkout
         uses: actions/checkout@v4
       - name: Setup pnpm
         uses: pnpm/action-setup@v2
         with:
-          version: 8
+          version: '8.12.1'
       - name: Install dependencies
         run: 'pnpm install --prefer-frozen-lockfile'
       - name: Typescript

.github/workflows/pypi-release.yml

@@ -1,13 +1,15 @@
 name: PyPI Release
 
 on:
-  push:
-    paths:
-      - 'invokeai/version/invokeai_version.py'
   workflow_dispatch:
+    inputs:
+      publish_package:
+        description: 'Publish build on PyPi? [true/false]'
+        required: true
+        default: 'false'
 
 jobs:
-  release:
+  build-and-release:
     if: github.repository == 'invoke-ai/InvokeAI'
     runs-on: ubuntu-22.04
     env:
@@ -15,19 +17,43 @@ jobs:
       TWINE_PASSWORD: ${{ secrets.PYPI_API_TOKEN }}
       TWINE_NON_INTERACTIVE: 1
     steps:
-      - name: checkout sources
-        uses: actions/checkout@v3
+      - name: Checkout
+        uses: actions/checkout@v4
 
-      - name: install deps
+      - 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 package
+      - name: Build python package
         run: python3 -m build
 
-      - name: check distribution
+      - 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
+      - name: Check PyPI versions
         if: github.ref == 'refs/heads/main' || startsWith(github.ref, 'refs/heads/release/')
         run: |
           pip install --upgrade requests
@@ -36,6 +62,6 @@ jobs:
           EXISTS=scripts.pypi_helper.local_on_pypi(); \
           print(f'PACKAGE_EXISTS={EXISTS}')" >> $GITHUB_ENV
 
-      - name: upload package
-        if: env.PACKAGE_EXISTS == 'False' && env.TWINE_PASSWORD != ''
+      - 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/test-invoke-pip.yml

@@ -58,7 +58,7 @@ jobs:
 
       - name: Check for changed python files
         id: changed-files
-        uses: tj-actions/changed-files@v37
+        uses: tj-actions/changed-files@v41
         with:
           files_yaml: |
             python:

README.md

@@ -1,10 +1,10 @@
 <div align="center">
 
-![project hero](https://github.com/invoke-ai/InvokeAI/assets/31807370/1a917d94-e099-4fa1-a70f-7dd8d0691018)
+![project hero](https://github.com/invoke-ai/InvokeAI/assets/31807370/6e3728c7-e90e-4711-905c-3b55844ff5be)
 
-# Invoke AI - Generative AI for Professional Creatives
-## Professional Creative Tools for Stable Diffusion, Custom-Trained Models, and more. 
-  To learn more about Invoke AI, get started instantly, or implement our Business solutions, visit [invoke.ai](https://invoke.ai)
+# Invoke - Professional Creative AI Tools for Visual Media 
+##  To learn more about Invoke, or implement our Business solutions, visit [invoke.com](https://www.invoke.com/about)
+  
 
 
 [![discord badge]][discord link]
@@ -56,7 +56,9 @@ the foundation for multiple commercial products.
 
 <div align="center">
 
-![canvas preview](https://github.com/invoke-ai/InvokeAI/raw/main/docs/assets/canvas_preview.png)
+
+![Highlighted Features - Canvas and Workflows](https://github.com/invoke-ai/InvokeAI/assets/31807370/708f7a82-084f-4860-bfbe-e2588c53548d)
+
 
 </div>
 
@@ -167,7 +169,7 @@ the command `npm install -g pnpm` if needed)
     _For Linux with an AMD GPU:_
 
     ```sh
-    pip install InvokeAI --use-pep517 --extra-index-url https://download.pytorch.org/whl/rocm5.4.2
+    pip install InvokeAI --use-pep517 --extra-index-url https://download.pytorch.org/whl/rocm5.6
     ```
 
     _For non-GPU systems:_
@@ -270,7 +272,7 @@ upgrade script.** See the next section for a Windows recipe.
 3. Select option [1] to upgrade to the latest release.
 
 4. Once the upgrade is finished you will be returned to the launcher
-menu. Select option [7] "Re-run the configure script to fix a broken
+menu. Select option [6] "Re-run the configure script to fix a broken
 install or to complete a major upgrade".
 
 This will run the configure script against the v2.3 directory and

docker/.env.sample

@@ -2,14 +2,17 @@
 ## Any environment variables supported by InvokeAI can be specified here,
 ## in addition to the examples below.
 
-# INVOKEAI_ROOT is the path to a path on the local filesystem where InvokeAI will store data.
+# HOST_INVOKEAI_ROOT is the path on the docker host's filesystem where InvokeAI will store data.
 # Outputs will also be stored here by default.
-# This **must** be an absolute path.
-INVOKEAI_ROOT=
+# If relative, it will be relative to the docker directory in which the docker-compose.yml file is located
+#HOST_INVOKEAI_ROOT=../../invokeai-data
+
+# INVOKEAI_ROOT is the path to the root of the InvokeAI repository within the container.
+# INVOKEAI_ROOT=~/invokeai
 
 # Get this value from your HuggingFace account settings page.
 # HUGGING_FACE_HUB_TOKEN=
 
 ## optional variables specific to the docker setup.
-# GPU_DRIVER=cuda # or rocm
+# GPU_DRIVER=nvidia #| rocm
 # CONTAINER_UID=1000

docker/Dockerfile

@@ -59,14 +59,16 @@ RUN --mount=type=cache,target=/root/.cache/pip \
 
 # #### Build the Web UI ------------------------------------
 
-FROM node:18 AS web-builder
+FROM node:20-slim AS web-builder
+ENV PNPM_HOME="/pnpm"
+ENV PATH="$PNPM_HOME:$PATH"
+RUN corepack enable
+
 WORKDIR /build
 COPY invokeai/frontend/web/ ./
-RUN --mount=type=cache,target=/usr/lib/node_modules \
-    npm install --include dev
-RUN --mount=type=cache,target=/usr/lib/node_modules \
-    yarn vite build
-
+RUN --mount=type=cache,target=/pnpm/store \
+    pnpm install --frozen-lockfile
+RUN npx vite build
 
 #### Runtime stage ---------------------------------------
 

docker/README.md

@@ -1,6 +1,14 @@
 # InvokeAI Containerized
 
-All commands are to be run from the `docker` directory: `cd docker`
+All commands should be run within the `docker` directory: `cd docker`
+
+## Quickstart :rocket:
+
+On a known working Linux+Docker+CUDA (Nvidia) system, execute `./run.sh` in this directory. It will take a few minutes - depending on your internet speed - to install the core models. Once the application starts up, open `http://localhost:9090` in your browser to Invoke!
+
+For more configuration options (using an AMD GPU, custom root directory location, etc): read on.
+
+## Detailed setup
 
 #### Linux
 
@@ -18,12 +26,12 @@ All commands are to be run from the `docker` directory: `cd docker`
 
 This is done via Docker Desktop preferences
 
-## Quickstart
+### Configure Invoke environment
 
 1. Make a copy of `env.sample` and name it `.env` (`cp env.sample .env` (Mac/Linux) or `copy example.env .env` (Windows)). Make changes as necessary. Set `INVOKEAI_ROOT` to an absolute path to:
     a. the desired location of the InvokeAI runtime directory, or
     b. an existing, v3.0.0 compatible runtime directory.
-1. `docker compose up`
+1. Execute `run.sh`
 
 The image will be built automatically if needed.
 
@@ -37,19 +45,21 @@ The runtime directory (holding models and outputs) will be created in the locati
 
 The Docker daemon on the system must be already set up to use the GPU. In case of Linux, this involves installing `nvidia-docker-runtime` and configuring the `nvidia` runtime as default. Steps will be different for AMD. Please see Docker documentation for the most up-to-date instructions for using your GPU with Docker.
 
+To use an AMD GPU, set `GPU_DRIVER=rocm` in your `.env` file.
+
 ## Customize
 
-Check the `.env.sample` file. It contains some environment variables for running in Docker. Copy it, name it `.env`, and fill it in with your own values. Next time you run `docker compose up`, your custom values will be used.
+Check the `.env.sample` file. It contains some environment variables for running in Docker. Copy it, name it `.env`, and fill it in with your own values. Next time you run `run.sh`, your custom values will be used.
 
 You can also set these values in `docker-compose.yml` directly, but `.env` will help avoid conflicts when code is updated.
 
-Example (values are optional, but setting `INVOKEAI_ROOT` is highly recommended):
+Values are optional, but setting `INVOKEAI_ROOT` is highly recommended. The default is `~/invokeai`. Example:
 
 ```bash
 INVOKEAI_ROOT=/Volumes/WorkDrive/invokeai
 HUGGINGFACE_TOKEN=the_actual_token
 CONTAINER_UID=1000
-GPU_DRIVER=cuda
+GPU_DRIVER=nvidia
 ```
 
 Any environment variables supported by InvokeAI can be set here - please see the [Configuration docs](https://invoke-ai.github.io/InvokeAI/features/CONFIGURATION/) for further detail.

docker/build.sh

@@ -1,11 +0,0 @@
-#!/usr/bin/env bash
-set -e
-
-build_args=""
-
-[[ -f ".env" ]] && build_args=$(awk '$1 ~ /\=[^$]/ {print "--build-arg " $0 " "}' .env)
-
-echo "docker compose build args:"
-echo $build_args
-
-docker compose build $build_args

docker/docker-compose.yml

@@ -2,23 +2,8 @@
 
 version: '3.8'
 
-services:
-  invokeai:
+x-invokeai: &invokeai
     image: "local/invokeai:latest"
-    # edit below to run on a container runtime other than nvidia-container-runtime.
-    # not yet tested with rocm/AMD GPUs
-    # Comment out the "deploy" section to run on CPU only
-    deploy:
-      resources:
-        reservations:
-          devices:
-            - driver: nvidia
-              count: 1
-              capabilities: [gpu]
-    # For AMD support, comment out the deploy section above and uncomment the devices section below:
-    #devices:
-    #  - /dev/kfd:/dev/kfd
-    #  - /dev/dri:/dev/dri
     build:
       context: ..
       dockerfile: docker/Dockerfile
@@ -36,7 +21,9 @@ services:
     ports:
       - "${INVOKEAI_PORT:-9090}:9090"
     volumes:
-      - ${INVOKEAI_ROOT:-~/invokeai}:${INVOKEAI_ROOT:-/invokeai}
+      - type: bind
+        source: ${HOST_INVOKEAI_ROOT:-${INVOKEAI_ROOT:-~/invokeai}}
+        target: ${INVOKEAI_ROOT:-/invokeai}
       - ${HF_HOME:-~/.cache/huggingface}:${HF_HOME:-/invokeai/.cache/huggingface}
       # - ${INVOKEAI_MODELS_DIR:-${INVOKEAI_ROOT:-/invokeai/models}}
       # - ${INVOKEAI_MODELS_CONFIG_PATH:-${INVOKEAI_ROOT:-/invokeai/configs/models.yaml}}
@@ -50,3 +37,27 @@ services:
     #   - |
     #     invokeai-model-install --yes --default-only --config_file ${INVOKEAI_ROOT}/config_custom.yaml
     #     invokeai-nodes-web --host 0.0.0.0
+
+services:
+  invokeai-nvidia:
+    <<: *invokeai
+    deploy:
+      resources:
+        reservations:
+          devices:
+            - driver: nvidia
+              count: 1
+              capabilities: [gpu]
+
+  invokeai-cpu:
+    <<: *invokeai
+    profiles:
+      - cpu
+
+  invokeai-rocm:
+    <<: *invokeai
+    devices:
+      - /dev/kfd:/dev/kfd
+      - /dev/dri:/dev/dri
+    profiles:
+      - rocm

docker/run.sh

@@ -1,11 +1,32 @@
 #!/usr/bin/env bash
-set -e
+set -e -o pipefail
 
-# This script is provided for backwards compatibility with the old docker setup.
-# it doesn't do much aside from wrapping the usual docker compose CLI.
+run() {
+  local scriptdir=$(dirname "${BASH_SOURCE[0]}")
+  cd "$scriptdir" || exit 1
 
-SCRIPTDIR=$(dirname "${BASH_SOURCE[0]}")
-cd "$SCRIPTDIR" || exit 1
+  local build_args=""
+  local profile=""
 
-docker compose up -d
-docker compose logs -f
+  touch .env
+  build_args=$(awk '$1 ~ /=[^$]/ && $0 !~ /^#/ {print "--build-arg " $0 " "}' .env) &&
+  profile="$(awk -F '=' '/GPU_DRIVER/ {print $2}' .env)"
+
+  [[ -z "$profile" ]] && profile="nvidia"
+
+  local service_name="invokeai-$profile"
+
+  if [[ ! -z "$build_args" ]]; then
+    printf "%s\n" "docker compose build args:"
+    printf "%s\n" "$build_args"
+  fi
+
+  docker compose build $build_args
+  unset build_args
+
+  printf "%s\n" "starting service $service_name"
+  docker compose --profile "$profile" up -d "$service_name"
+  docker compose logs -f
+}
+
+run

docs/contributing/DOWNLOAD_QUEUE.md

@@ -0,0 +1,277 @@
+# The InvokeAI Download Queue
+
+The DownloadQueueService provides a multithreaded parallel download
+queue for arbitrary URLs, with queue prioritization, event handling,
+and restart capabilities.
+
+## Simple Example
+
+```
+from invokeai.app.services.download import DownloadQueueService, TqdmProgress
+
+download_queue = DownloadQueueService()
+for url in ['https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/assets/a-painting-of-a-fire.png?raw=true',
+            'https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/assets/birdhouse.png?raw=true',
+            'https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/assets/missing.png',
+            'https://civitai.com/api/download/models/152309?type=Model&format=SafeTensor',
+            ]:
+
+    # urls start downloading as soon as download() is called
+    download_queue.download(source=url,
+                            dest='/tmp/downloads',
+                            on_progress=TqdmProgress().update
+                            )
+
+download_queue.join()  # wait for all downloads to finish
+for job in download_queue.list_jobs():
+    print(job.model_dump_json(exclude_none=True, indent=4),"\n")
+```
+
+Output:
+
+```
+{
+    "source": "https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/assets/a-painting-of-a-fire.png?raw=true",
+    "dest": "/tmp/downloads",
+    "id": 0,
+    "priority": 10,
+    "status": "completed",
+    "download_path": "/tmp/downloads/a-painting-of-a-fire.png",
+    "job_started": "2023-12-04T05:34:41.742174",
+    "job_ended": "2023-12-04T05:34:42.592035",
+    "bytes": 666734,
+    "total_bytes": 666734
+} 
+
+{
+    "source": "https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/assets/birdhouse.png?raw=true",
+    "dest": "/tmp/downloads",
+    "id": 1,
+    "priority": 10,
+    "status": "completed",
+    "download_path": "/tmp/downloads/birdhouse.png",
+    "job_started": "2023-12-04T05:34:41.741975",
+    "job_ended": "2023-12-04T05:34:42.652841",
+    "bytes": 774949,
+    "total_bytes": 774949
+}
+
+{
+    "source": "https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/assets/missing.png",
+    "dest": "/tmp/downloads",
+    "id": 2,
+    "priority": 10,
+    "status": "error",
+    "job_started": "2023-12-04T05:34:41.742079",
+    "job_ended": "2023-12-04T05:34:42.147625",
+    "bytes": 0,
+    "total_bytes": 0,
+    "error_type": "HTTPError(Not Found)",
+    "error": "Traceback (most recent call last):\n  File \"/home/lstein/Projects/InvokeAI/invokeai/app/services/download/download_default.py\", line 182, in _download_next_item\n    self._do_download(job)\n  File \"/home/lstein/Projects/InvokeAI/invokeai/app/services/download/download_default.py\", line 206, in _do_download\n    raise HTTPError(resp.reason)\nrequests.exceptions.HTTPError: Not Found\n"
+}
+
+{
+    "source": "https://civitai.com/api/download/models/152309?type=Model&format=SafeTensor",
+    "dest": "/tmp/downloads",
+    "id": 3,
+    "priority": 10,
+    "status": "completed",
+    "download_path": "/tmp/downloads/xl_more_art-full_v1.safetensors",
+    "job_started": "2023-12-04T05:34:42.147645",
+    "job_ended": "2023-12-04T05:34:43.735990",
+    "bytes": 719020768,
+    "total_bytes": 719020768
+} 
+```
+
+##  The API
+
+The default download queue is `DownloadQueueService`, an
+implementation of ABC `DownloadQueueServiceBase`. It juggles multiple
+background download requests and provides facilities for interrogating
+and cancelling the requests. Access to a current or past download task
+is mediated via `DownloadJob` objects which report the current status
+of a job request
+
+### The Queue Object
+
+A default download queue is located in
+`ApiDependencies.invoker.services.download_queue`. However, you can
+create additional instances if you need to isolate your queue from the
+main one.
+
+```
+queue = DownloadQueueService(event_bus=events)
+```
+
+`DownloadQueueService()` takes three optional arguments:
+
+| **Argument** | **Type**          |  **Default**  | **Description** |
+|----------------|-----------------|---------------|-----------------|
+| `max_parallel_dl`  | int                         | 5    | Maximum number of simultaneous downloads allowed |
+| `event_bus` | EventServiceBase   | None | System-wide FastAPI event bus for reporting download events |
+| `requests_session` | requests.sessions.Session   | None | An alternative requests Session object to use for the download |
+
+`max_parallel_dl` specifies how many download jobs are allowed to run
+simultaneously. Each will run in a different thread of execution.
+
+`event_bus` is an EventServiceBase, typically the one created at
+InvokeAI startup. If present, download events are periodically emitted
+on this bus to allow clients to follow download progress.
+
+`requests_session` is a url library requests Session object. It is
+used for testing.
+
+### The Job object
+
+The queue operates on a series of download job objects. These objects
+specify the source and destination of the download, and keep track of
+the progress of the download.
+
+The only job type currently implemented is `DownloadJob`, a pydantic object with the
+following fields:
+
+| **Field**      | **Type**        |  **Default**  | **Description** |
+|----------------|-----------------|---------------|-----------------|
+| _Fields passed in at job creation time_                               |
+| `source`         | AnyHttpUrl      |               | Where to download from |
+| `dest`           | Path            |               | Where to download to              |
+| `access_token`   | str             |               | [optional] string containing authentication token for access |
+| `on_start`       | Callable        |               | [optional] callback when the download starts |
+| `on_progress`    | Callable        |               | [optional] callback called at intervals during download progress |
+| `on_complete`    | Callable        |               | [optional] callback called after successful download completion |
+| `on_error`       | Callable        |               | [optional] callback called after an error occurs  |
+| `id`             | int             | auto assigned | Job ID, an integer >= 0           |
+| `priority`       | int             | 10            | Job priority. Lower priorities run before higher priorities |
+|                                                                                                        |
+| _Fields updated over the course of the download task_
+| `status`         | DownloadJobStatus|              | Status code                                |
+| `download_path`  | Path |              | Path to the location of the downloaded file |
+| `job_started`    | float            |              | Timestamp for when the job started running |
+| `job_ended`      | float            |              | Timestamp for when the job completed or errored out |
+| `job_sequence`   | int              |              | A counter that is incremented each time a model is dequeued |
+| `bytes`          | int              | 0            | Bytes downloaded so far   |
+| `total_bytes`    | int              | 0            | Total size of the file at the remote site  |
+| `error_type`     | str              |              | String version of the exception that caused an error during download |
+| `error`          | str              |              | String version of the traceback associated with an error |
+| `cancelled`      | bool             | False        | Set to true if the job was cancelled by the caller|
+
+When you create a job, you can assign it a `priority`. If multiple
+jobs are queued, the job with the lowest priority runs first.
+
+Every job has a `source` and a `dest`. `source` is a pydantic.networks AnyHttpUrl object.
+The `dest` is a path on the local filesystem that specifies the
+destination for the downloaded object. Its semantics are
+described below.
+
+When the job is submitted, it is assigned a numeric `id`. The id can
+then be used to fetch the job object from the queue.
+
+The `status` field is updated by the queue to indicate where the job
+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 ** |
+|--------------|---------------------|-------------------|
+| `WAITING`      | waiting           | Job is on the queue but not yet running|
+| `RUNNING`      | running           | The download is started                |
+| `COMPLETED`    | completed         | Job has finished its work without an error |
+| `ERROR`        | error             | Job encountered an error and will not run again|
+
+`job_started` and `job_ended` indicate when the job
+was started (using a python timestamp) and when it completed.
+
+In case of an error, the job's status will be set to `DownloadJobStatus.ERROR`, the text of the
+Exception that caused the error will be placed in the `error_type`
+field and the traceback that led to the error will be in `error`.
+
+A cancelled job will have status `DownloadJobStatus.ERROR` and an
+`error_type` field of "DownloadJobCancelledException". In addition,
+the job's `cancelled` property will be set to True.
+
+### Callbacks
+
+Download jobs can be associated with a series of callbacks, each with
+the signature `Callable[["DownloadJob"], None]`. The callbacks are assigned
+using optional arguments `on_start`, `on_progress`, `on_complete` and
+`on_error`. When the corresponding event occurs, the callback wil be
+invoked and passed the job. The callback will be run in a `try:`
+context in the same thread as the download job. Any exceptions that
+occur during execution of the callback will be caught and converted
+into a log error message, thereby allowing the download to continue.
+
+#### `TqdmProgress`
+
+The `invokeai.app.services.download.download_default` module defines a
+class named `TqdmProgress` which can be used as an `on_progress`
+handler to display a completion bar in the console. Use as follows:
+
+```
+from invokeai.app.services.download import TqdmProgress
+
+download_queue.download(source='http://some.server.somewhere/some_file',
+                        dest='/tmp/downloads',
+                        on_progress=TqdmProgress().update
+                        )
+
+```
+
+### Events
+
+If the queue was initialized with the InvokeAI event bus (the case
+when using `ApiDependencies.invoker.services.download_queue`), then
+download events will also be issued on the bus. The events are:
+
+* `download_started` -- This is issued when a job is taken off the
+queue and a request is made to the remote server for the URL headers, but before any data
+has been downloaded. The event payload will contain the keys `source`
+and `download_path`. The latter contains the path that the URL will be
+downloaded to.
+
+* `download_progress -- This is issued periodically as the download
+runs. The payload contains the keys `source`, `download_path`,
+`current_bytes` and `total_bytes`. The latter two fields can be
+used to display the percent complete.
+
+* `download_complete` -- This is issued when the download completes
+successfully. The payload contains the keys `source`, `download_path`
+and `total_bytes`.
+
+* `download_error` -- This is issued when the download stops because
+of an error condition. The payload contains the fields `error_type`
+and `error`. The former is the text representation of the exception,
+and the latter is a traceback showing where the error occurred.
+
+### Job control
+
+To create a job call the queue's `download()` method. You can list all
+jobs using `list_jobs()`, fetch a single job by its with
+`id_to_job()`, cancel a running job with `cancel_job()`, cancel all
+running jobs with `cancel_all_jobs()`, and wait for all jobs to finish
+with `join()`.
+
+#### job = queue.download(source, dest, priority, access_token)
+
+Create a new download job and put it on the queue, returning the
+DownloadJob object.
+
+#### jobs = queue.list_jobs()
+
+Return a list of all active and inactive `DownloadJob`s.
+
+#### job = queue.id_to_job(id)
+
+Return the job corresponding to given ID.
+
+Return a list of all active and inactive `DownloadJob`s.
+
+#### queue.prune_jobs()
+
+Remove inactive (complete or errored) jobs from the listing returned
+by `list_jobs()`.
+
+#### queue.join()
+
+Block until all pending jobs have run to completion or errored out.
+

docs/contributing/contribution_guides/contributingToFrontend.md

@@ -1,75 +0,0 @@
-# Contributing to the Frontend
-
-# InvokeAI Web UI
-
-- [InvokeAI Web UI](https://github.com/invoke-ai/InvokeAI/tree/main/invokeai/frontend/web/docs#invokeai-web-ui)
-    - [Stack](https://github.com/invoke-ai/InvokeAI/tree/main/invokeai/frontend/web/docs#stack)
-    - [Contributing](https://github.com/invoke-ai/InvokeAI/tree/main/invokeai/frontend/web/docs#contributing)
-        - [Dev Environment](https://github.com/invoke-ai/InvokeAI/tree/main/invokeai/frontend/web/docs#dev-environment)
-        - [Production builds](https://github.com/invoke-ai/InvokeAI/tree/main/invokeai/frontend/web/docs#production-builds)
-
-The UI is a fairly straightforward Typescript React app, with the Unified Canvas being more complex.
-
-Code is located in `invokeai/frontend/web/` for review.
-
-## Stack
-
-State management is Redux via [Redux Toolkit](https://github.com/reduxjs/redux-toolkit). We lean heavily on RTK:
-
-- `createAsyncThunk` for HTTP requests
-- `createEntityAdapter` for fetching images and models
-- `createListenerMiddleware` for workflows
-
-The API client and associated types are generated from the OpenAPI schema. See API_CLIENT.md.
-
-Communication with server is a mix of HTTP and [socket.io](https://github.com/socketio/socket.io-client) (with a simple socket.io redux middleware to help).
-
-[Chakra-UI](https://github.com/chakra-ui/chakra-ui) & [Mantine](https://github.com/mantinedev/mantine) for components and styling.
-
-[Konva](https://github.com/konvajs/react-konva) for the canvas, but we are pushing the limits of what is feasible with it (and HTML canvas in general). We plan to rebuild it with [PixiJS](https://github.com/pixijs/pixijs) to take advantage of WebGL's improved raster handling.
-
-[Vite](https://vitejs.dev/) for bundling.
-
-Localisation is via [i18next](https://github.com/i18next/react-i18next), but translation happens on our [Weblate](https://hosted.weblate.org/engage/invokeai/) project. Only the English source strings should be changed on this repo.
-
-## Contributing
-
-Thanks for your interest in contributing to the InvokeAI Web UI!
-
-We encourage you to ping @psychedelicious and @blessedcoolant on [Discord](https://discord.gg/ZmtBAhwWhy) if you want to contribute, just to touch base and ensure your work doesn't conflict with anything else going on. The project is very active.
-
-### Dev Environment
-
-**Setup** 
-
-1. Install [node](https://nodejs.org/en/download/). You can confirm node is installed with:
-```bash
-node --version
-```
-2. Install [yarn classic](https://classic.yarnpkg.com/lang/en/) and confirm it is installed by running this:
-```bash
-npm install --global yarn
-yarn --version
-```
-
-From `invokeai/frontend/web/` run `yarn install` to get everything set up.
-
-Start everything in dev mode:
-1. Ensure your virtual environment is running
-2. Start the dev server: `yarn dev`
-3. Start the InvokeAI Nodes backend: `python scripts/invokeai-web.py # run from the repo root`
-4. Point your browser to the dev server address e.g. [http://localhost:5173/](http://localhost:5173/)
-
-### VSCode Remote Dev
-
-We've noticed an intermittent issue with the VSCode Remote Dev port forwarding. If you use this feature of VSCode, you may intermittently click the Invoke button and then get nothing until the request times out. Suggest disabling the IDE's port forwarding feature and doing it manually via SSH:
-
-`ssh -L 9090:localhost:9090 -L 5173:localhost:5173 user@host`
-
-### Production builds
-
-For a number of technical and logistical reasons, we need to commit UI build artefacts to the repo.
-
-If you submit a PR, there is a good chance we will ask you to include a separate commit with a build of the app.
-
-To build for production, run `yarn build`.

docs/contributing/contribution_guides/development.md

@@ -12,7 +12,7 @@ To get started, take a look at our [new contributors checklist](newContributorCh
 Once you're setup, for more information, you can review the documentation specific to your area of interest:
 
 * #### [InvokeAI Architecure](../ARCHITECTURE.md)
-* #### [Frontend Documentation](./contributingToFrontend.md)
+* #### [Frontend Documentation](https://github.com/invoke-ai/InvokeAI/tree/main/invokeai/frontend/web)
 * #### [Node Documentation](../INVOCATIONS.md)
 * #### [Local Development](../LOCAL_DEVELOPMENT.md)
 

docs/deprecated/2to3.md

@@ -0,0 +1,53 @@
+## :octicons-log-16: Important Changes Since Version 2.3
+
+### Nodes
+
+Behind the scenes, InvokeAI has been completely rewritten to support
+"nodes," small unitary operations that can be combined into graphs to
+form arbitrary workflows. For example, there is a prompt node that
+processes the prompt string and feeds it to a text2latent node that
+generates a latent image. The latents are then fed to a latent2image
+node that translates the latent image into a PNG.
+
+The WebGUI has a node editor that allows you to graphically design and
+execute custom node graphs. The ability to save and load graphs is
+still a work in progress, but coming soon.
+
+### Command-Line Interface Retired
+
+All "invokeai" command-line interfaces have been retired as of version
+3.4.
+
+To launch the Web GUI from the command-line, use the command
+`invokeai-web` rather than the traditional `invokeai --web`.
+
+### ControlNet
+
+This version of InvokeAI features ControlNet, a system that allows you
+to achieve exact poses for human and animal figures by providing a
+model to follow. Full details are found in [ControlNet](features/CONTROLNET.md)
+
+### New Schedulers
+
+The list of schedulers has been completely revamped and brought up to date:
+
+| **Short Name** | **Scheduler**                   | **Notes**                   |
+|----------------|---------------------------------|-----------------------------|
+| **ddim**       | DDIMScheduler                   |                             |
+| **ddpm**       | DDPMScheduler                   |                             |
+| **deis**       | DEISMultistepScheduler          |                             |
+| **lms**        | LMSDiscreteScheduler            |                             |
+| **pndm**       | PNDMScheduler                   |                             |
+| **heun**       | HeunDiscreteScheduler           | original noise schedule     |
+| **heun_k**     | HeunDiscreteScheduler           | using karras noise schedule |
+| **euler**      | EulerDiscreteScheduler          | original noise schedule     |
+| **euler_k**    | EulerDiscreteScheduler          | using karras noise schedule |
+| **kdpm_2**     | KDPM2DiscreteScheduler          |                             |
+| **kdpm_2_a**   | KDPM2AncestralDiscreteScheduler |                             |
+| **dpmpp_2s**   | DPMSolverSinglestepScheduler    |                             |
+| **dpmpp_2m**   | DPMSolverMultistepScheduler     | original noise scnedule     |
+| **dpmpp_2m_k** | DPMSolverMultistepScheduler     | using karras noise schedule |
+| **unipc**      | UniPCMultistepScheduler         | CPU only                    |
+| **lcm**        | LCMScheduler                    |                             |
+
+Please see [3.0.0 Release Notes](https://github.com/invoke-ai/InvokeAI/releases/tag/v3.0.0) for further details.

docs/features/UNIFIED_CANVAS.md

@@ -229,29 +229,28 @@ clarity on the intent and common use cases we expect for utilizing them.
   currently being rendered by your browser into a merged copy of the image. This
   lowers the resource requirements and should improve performance.
 
-### Seam Correction
+### Compositing / Seam Correction
 
 When doing Inpainting or Outpainting, Invoke needs to merge the pixels generated
-by Stable Diffusion into your existing image. To do this, the area around the
-`seam` at the boundary between your image and the new generation is
+by Stable Diffusion into your existing image. This is achieved through compositing - the area around the the boundary between your image and the new generation is
 automatically blended to produce a seamless output. In a fully automatic
-process, a mask is generated to cover the seam, and then the area of the seam is
+process, a mask is generated to cover the boundary, and then the area of the boundary is
 Inpainted.
 
 Although the default options should work well most of the time, sometimes it can
-help to alter the parameters that control the seam Inpainting. A wider seam and
-a blur setting of about 1/3 of the seam have been noted as producing
-consistently strong results (e.g. 96 wide and 16 blur - adds up to 32 blur with
-both sides). Seam strength of 0.7 is best for reducing hard seams.
+help to alter the parameters that control the Compositing. A larger blur and
+a blur setting  have been noted as producing
+consistently strong results . Strength of 0.7 is best for reducing hard seams.
+
+- **Mode** - What part of the image will have the the Compositing applied to it.
+  - **Mask edge** will apply Compositing to the edge of the masked area
+  - **Mask** will apply Compositing to the entire masked area
+  - **Unmasked** will apply Compositing to the entire image
+- **Steps** - Number of generation steps that will occur during the Coherence Pass, similar to Denoising Steps. Higher step counts will generally have better results.
+- **Strength** - How much noise is added for the Coherence Pass, similar to Denoising Strength. A strength of 0 will result in an unchanged image, while a strength of 1 will result in an image with a completely new area as defined by the Mode setting.
+- **Blur** - Adjusts the pixel radius of the the mask. A larger blur radius will cause the mask to extend past the visibly masked area, while too small of a blur radius will result in a mask that is smaller than the visibly masked area.
+- **Blur Method** - The method of blur applied to the masked area. 
 
-- **Seam Size** - The size of the seam masked area. Set higher to make a larger
-  mask around the seam.
-- **Seam Blur** - The size of the blur that is applied on _each_ side of the
-  masked area.
-- **Seam Strength** - The Image To Image Strength parameter used for the
-  Inpainting generation that is applied to the seam area.
-- **Seam Steps** - The number of generation steps that should be used to Inpaint
-  the seam.
 
 ### Infill & Scaling
 

docs/index.md

@@ -18,7 +18,7 @@ title: Home
       width: 100%;
       max-width: 100%;
       height: 50px;
-      background-color: #448AFF;
+      background-color: #35A4DB;
       color: #fff;
       font-size: 16px;
       border: none;
@@ -43,7 +43,7 @@ title: Home
 <div align="center" markdown>
 
 
-[![project logo](assets/invoke_ai_banner.png)](https://github.com/invoke-ai/InvokeAI)
+[![project logo](https://github.com/invoke-ai/InvokeAI/assets/31807370/6e3728c7-e90e-4711-905c-3b55844ff5be)](https://github.com/invoke-ai/InvokeAI)
 
 [![discord badge]][discord link]
 
@@ -117,6 +117,11 @@ Mac and Linux machines, and runs on GPU cards with as little as 4 GB of RAM.
 
 ## :octicons-gift-24: InvokeAI Features
 
+### Installation
+  - [Automated Installer](installation/010_INSTALL_AUTOMATED.md)
+  - [Manual Installation](installation/020_INSTALL_MANUAL.md)
+  - [Docker Installation](installation/040_INSTALL_DOCKER.md)
+
 ### The InvokeAI Web Interface
 - [WebUI overview](features/WEB.md)
 - [WebUI hotkey reference guide](features/WEBUIHOTKEYS.md)
@@ -145,60 +150,6 @@ Mac and Linux machines, and runs on GPU cards with as little as 4 GB of RAM.
 - [Guide to InvokeAI Runtime Settings](features/CONFIGURATION.md)
 - [Database Maintenance and other Command Line Utilities](features/UTILITIES.md)
 
-## :octicons-log-16: Important Changes Since Version 2.3
-
-### Nodes
-
-Behind the scenes, InvokeAI has been completely rewritten to support
-"nodes," small unitary operations that can be combined into graphs to
-form arbitrary workflows. For example, there is a prompt node that
-processes the prompt string and feeds it to a text2latent node that
-generates a latent image. The latents are then fed to a latent2image
-node that translates the latent image into a PNG.
-
-The WebGUI has a node editor that allows you to graphically design and
-execute custom node graphs. The ability to save and load graphs is
-still a work in progress, but coming soon.
-
-### Command-Line Interface Retired
-
-All "invokeai" command-line interfaces have been retired as of version
-3.4.
-
-To launch the Web GUI from the command-line, use the command
-`invokeai-web` rather than the traditional `invokeai --web`.
-
-### ControlNet
-
-This version of InvokeAI features ControlNet, a system that allows you
-to achieve exact poses for human and animal figures by providing a
-model to follow. Full details are found in [ControlNet](features/CONTROLNET.md)
-
-### New Schedulers
-
-The list of schedulers has been completely revamped and brought up to date:
-
-| **Short Name** | **Scheduler**                   | **Notes**                   |
-|----------------|---------------------------------|-----------------------------|
-| **ddim**       | DDIMScheduler                   |                             |
-| **ddpm**       | DDPMScheduler                   |                             |
-| **deis**       | DEISMultistepScheduler          |                             |
-| **lms**        | LMSDiscreteScheduler            |                             |
-| **pndm**       | PNDMScheduler                   |                             |
-| **heun**       | HeunDiscreteScheduler           | original noise schedule     |
-| **heun_k**     | HeunDiscreteScheduler           | using karras noise schedule |
-| **euler**      | EulerDiscreteScheduler          | original noise schedule     |
-| **euler_k**    | EulerDiscreteScheduler          | using karras noise schedule |
-| **kdpm_2**     | KDPM2DiscreteScheduler          |                             |
-| **kdpm_2_a**   | KDPM2AncestralDiscreteScheduler |                             |
-| **dpmpp_2s**   | DPMSolverSinglestepScheduler    |                             |
-| **dpmpp_2m**   | DPMSolverMultistepScheduler     | original noise scnedule     |
-| **dpmpp_2m_k** | DPMSolverMultistepScheduler     | using karras noise schedule |
-| **unipc**      | UniPCMultistepScheduler         | CPU only                    |
-| **lcm**        | LCMScheduler                    |                             |
-
-Please see [3.0.0 Release Notes](https://github.com/invoke-ai/InvokeAI/releases/tag/v3.0.0) for further details.
-
 ## :material-target: Troubleshooting
 
 Please check out our **[:material-frequently-asked-questions:

docs/installation/010_INSTALL_AUTOMATED.md

@@ -477,7 +477,7 @@ Then type the following commands:
 
 === "AMD System"
     ```bash
-    pip install torch torchvision --force-reinstall --extra-index-url https://download.pytorch.org/whl/rocm5.4.2
+    pip install torch torchvision --force-reinstall --extra-index-url https://download.pytorch.org/whl/rocm5.6
     ```
 
 ### Corrupted configuration file

docs/installation/020_INSTALL_MANUAL.md

@@ -154,7 +154,7 @@ manager, please follow these steps:
     === "ROCm (AMD)"
 
         ```bash
-        pip install InvokeAI --use-pep517 --extra-index-url https://download.pytorch.org/whl/rocm5.4.2
+        pip install InvokeAI --use-pep517 --extra-index-url https://download.pytorch.org/whl/rocm5.6
         ```
 
     === "CPU (Intel Macs & non-GPU systems)"
@@ -313,7 +313,7 @@ code for InvokeAI. For this to work, you will need to install the
 on your system, please see the [Git Installation
 Guide](https://github.com/git-guides/install-git)
 
-You will also need to install the [frontend development toolchain](https://github.com/invoke-ai/InvokeAI/blob/main/docs/contributing/contribution_guides/contributingToFrontend.md).
+You will also need to install the [frontend development toolchain](https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/frontend/web/README.md).
 
 If you have a "normal" installation, you should create a totally separate virtual environment for the git-based installation, else the two may interfere.
 
@@ -345,7 +345,7 @@ installation protocol (important!)
 
     === "ROCm (AMD)"
         ```bash
-        pip install -e . --use-pep517 --extra-index-url https://download.pytorch.org/whl/rocm5.4.2
+        pip install -e . --use-pep517 --extra-index-url https://download.pytorch.org/whl/rocm5.6
         ```
 
     === "CPU (Intel Macs & non-GPU systems)"
@@ -361,7 +361,7 @@ installation protocol (important!)
     Be sure to pass `-e` (for an editable install) and don't forget the
     dot ("."). It is part of the command.
 
-5.  Install the [frontend toolchain](https://github.com/invoke-ai/InvokeAI/blob/main/docs/contributing/contribution_guides/contributingToFrontend.md) and do a production build of the UI as described.
+5.  Install the [frontend toolchain](https://github.com/invoke-ai/InvokeAI/blob/main/invokeai/frontend/web/README.md) and do a production build of the UI as described.
 
 6.  You can now run `invokeai` and its related commands. The code will be
     read from the repository, so that you can edit the .py source files

docs/installation/030_INSTALL_CUDA_AND_ROCM.md

@@ -134,7 +134,7 @@ recipes are available
 
 When installing torch and torchvision manually with `pip`, remember to provide
 the argument `--extra-index-url
-https://download.pytorch.org/whl/rocm5.4.2` as described in the [Manual
+https://download.pytorch.org/whl/rocm5.6` as described in the [Manual
 Installation Guide](020_INSTALL_MANUAL.md).
 
 This will be done automatically for you if you use the installer

docs/installation/INSTALLATION.md

@@ -18,13 +18,18 @@ either an Nvidia-based card (with CUDA support) or an AMD card (using the ROCm
 driver).
 
 
-## **[Automated Installer](010_INSTALL_AUTOMATED.md)**
-✅ This is the recommended installation method for first-time users. 
+## **[Automated Installer (Recommended)](010_INSTALL_AUTOMATED.md)**
+ ✅ This is the recommended installation method for first-time users. 
 
   This is a script that will install all of InvokeAI's essential
-  third party libraries and InvokeAI itself. It includes access to a
-  "developer console" which will help us debug problems with you and
-  give you to access experimental features.
+  third party libraries and InvokeAI itself.
+
+🖥️ **Download the latest installer .zip file here** : https://github.com/invoke-ai/InvokeAI/releases/latest
+  
+- *Look for the file labelled "InvokeAI-installer-v3.X.X.zip" at the bottom of the page*
+- If you experience issues, read through the full [installation instructions](010_INSTALL_AUTOMATED.md) to make sure you have met all of the installation requirements. If you need more help, join the [Discord](discord.gg/invoke-ai) or create an issue on [Github](https://github.com/invoke-ai/InvokeAI).
+
+
 
 ## **[Manual Installation](020_INSTALL_MANUAL.md)**
 This method is recommended for experienced users and developers.

docs/javascripts/init_kapa_widget.js

@@ -1,10 +1,10 @@
-document.addEventListener("DOMContentLoaded", function () {
-  var script = document.createElement("script");
-  script.src = "https://widget.kapa.ai/kapa-widget.bundle.js";
-  script.setAttribute("data-website-id", "b5973bb1-476b-451e-8cf4-98de86745a10");
-  script.setAttribute("data-project-name", "Invoke.AI");
-  script.setAttribute("data-project-color", "#11213C");
-  script.setAttribute("data-project-logo", "https://avatars.githubusercontent.com/u/113954515?s=280&v=4");
-  script.async = true;
-  document.head.appendChild(script);
-});
+document.addEventListener("DOMContentLoaded", function () {
+  var script = document.createElement("script");
+  script.src = "https://widget.kapa.ai/kapa-widget.bundle.js";
+  script.setAttribute("data-website-id", "b5973bb1-476b-451e-8cf4-98de86745a10");
+  script.setAttribute("data-project-name", "Invoke.AI");
+  script.setAttribute("data-project-color", "#11213C");
+  script.setAttribute("data-project-logo", "https://avatars.githubusercontent.com/u/113954515?s=280&v=4");
+  script.async = true;
+  document.head.appendChild(script);
+});

docs/nodes/NODES.md

@@ -6,10 +6,17 @@ If you're not familiar with Diffusion, take a look at our [Diffusion Overview.](
 
 ## Features
 
+### Workflow Library
+The Workflow Library enables you to save workflows to the Invoke database, allowing you to easily creating, modify and share workflows as needed. 
+
+A curated set of workflows are provided by default - these are designed to help explain important nodes' usage in the Workflow Editor.
+
+![workflow_library](../assets/nodes/workflow_library.png)
+
 ### Linear View
 The Workflow Editor allows you to create a UI for your workflow, to make it easier to iterate on your generations. 
 
-To add an input to the Linear UI, right click on the input label and select "Add to Linear View".
+To add an input to the Linear UI, right click on the **input label** and select "Add to Linear View".
 
 The Linear UI View will also be part of the saved workflow, allowing you share workflows and enable other to use them, regardless of complexity. 
 
@@ -30,7 +37,7 @@ Any node or input field can be renamed in the workflow editor. If the input fiel
 Nodes have a "Use Cache" option in their footer. This allows for performance improvements by using the previously cached values during the workflow processing. 
 
 
-## Important Concepts 
+## Important Nodes & Concepts 
 
 There are several node grouping concepts that can be examined with a narrow focus. These (and other) groupings can be pieced together to make up functional graph setups, and are important to understanding how groups of nodes work together as part of a whole. Note that the screenshots below aren't examples of complete functioning node graphs (see Examples).
 
@@ -56,7 +63,7 @@ The ImageToLatents node takes in a pixel image and a VAE and outputs a latents.
 
 It is common to want to use both the same seed (for continuity) and random seeds (for variety). To define a seed, simply enter it into the 'Seed' field on a noise node. Conversely, the RandomInt node generates a random integer between 'Low' and 'High', and can be used as input to the 'Seed' edge point on a noise node to randomize your seed.
 
-![groupsrandseed](../assets/nodes/groupsrandseed.png)
+![groupsrandseed](../assets/nodes/groupsnoise.png)
 
 ### ControlNet
 

docs/nodes/communityNodes.md

@@ -13,6 +13,8 @@ If you'd prefer, you can also just download the whole node folder from the linke
 To use a community workflow, download the the `.json` node graph file and load it into Invoke AI via the **Load Workflow** button in the Workflow Editor. 
 
 - Community Nodes
+    + [Adapters-Linked](#adapters-linked-nodes)
+    + [Autostereogram](#autostereogram-nodes)
     + [Average Images](#average-images)
     + [Clean Image Artifacts After Cut](#clean-image-artifacts-after-cut)
     + [Close Color Mask](#close-color-mask) 
@@ -24,7 +26,7 @@ To use a community workflow, download the the `.json` node graph file and load i
     + [GPT2RandomPromptMaker](#gpt2randompromptmaker)
     + [Grid to Gif](#grid-to-gif)
     + [Halftone](#halftone)
-    + [Ideal Size](#ideal-size)
+    + [Hand Refiner with MeshGraphormer](#hand-refiner-with-meshgraphormer)
     + [Image and Mask Composition Pack](#image-and-mask-composition-pack)
     + [Image Dominant Color](#image-dominant-color)
     + [Image to Character Art Image Nodes](#image-to-character-art-image-nodes)
@@ -32,12 +34,15 @@ To use a community workflow, download the the `.json` node graph file and load i
     + [Image Resize Plus](#image-resize-plus)
     + [Load Video Frame](#load-video-frame)
     + [Make 3D](#make-3d)
-    + [Mask Operations](#mask-operations)    
+    + [Mask Operations](#mask-operations)
     + [Match Histogram](#match-histogram)
-    + [Negative Image](#negative-image)    
+    + [Metadata-Linked](#metadata-linked-nodes)
+    + [Negative Image](#negative-image)
+    + [Nightmare Promptgen](#nightmare-promptgen)    
     + [Oobabooga](#oobabooga)
     + [Prompt Tools](#prompt-tools)
     + [Remote Image](#remote-image)
+    + [BriaAI Background Remove](#briaai-remove-background)
     + [Remove Background](#remove-background)    
     + [Retroize](#retroize)
     + [Size Stepper Nodes](#size-stepper-nodes)
@@ -51,6 +56,30 @@ To use a community workflow, download the the `.json` node graph file and load i
 - [Help](#help)
 
 
+--------------------------------
+### Adapters Linked Nodes
+
+**Description:** A set of nodes for linked adapters (ControlNet, IP-Adaptor & T2I-Adapter). This allows multiple adapters to be chained together without using a `collect` node which means it can be used inside an `iterate` node without any collecting on every iteration issues.
+
+- `ControlNet-Linked` - Collects ControlNet info to pass to other nodes.
+- `IP-Adapter-Linked` - Collects IP-Adapter info to pass to other nodes.
+- `T2I-Adapter-Linked` - Collects T2I-Adapter info to pass to other nodes.
+
+Note: These are inherited from the core nodes so any update to the core nodes should be reflected in these. 
+
+**Node Link:** https://github.com/skunkworxdark/adapters-linked-nodes
+
+--------------------------------
+### Autostereogram Nodes
+
+**Description:** Generate autostereogram images from a depth map. This is not a very practically useful node but more a 90s nostalgic indulgence as I used to love these images as a kid.
+
+**Node Link:** https://github.com/skunkworxdark/autostereogram_nodes
+
+**Example Usage:**
+</br>
+<img src="https://github.com/skunkworxdark/autostereogram_nodes/blob/main/images/spider.png" width="200" /> -> <img src="https://github.com/skunkworxdark/autostereogram_nodes/blob/main/images/spider-depth.png" width="200" /> -> <img src="https://github.com/skunkworxdark/autostereogram_nodes/raw/main/images/spider-dots.png" width="200" /> <img src="https://github.com/skunkworxdark/autostereogram_nodes/raw/main/images/spider-pattern.png" width="200" />
+
 --------------------------------
 ### Average Images
 
@@ -181,13 +210,18 @@ CMYK Halftone Output:
 <img src="https://github.com/invoke-ai/InvokeAI/assets/34005131/c59c578f-db8e-4d66-8c66-2851752d75ea" width="300" />
 
 --------------------------------
-### Ideal Size
 
-**Description:** This node calculates an ideal image size for a first pass of a multi-pass upscaling. The aim is to avoid duplication that results from choosing a size larger than the model is capable of.
+### Hand Refiner with MeshGraphormer
 
-**Node Link:** https://github.com/JPPhoto/ideal-size-node
+**Description**: Hand Refiner takes in your image and automatically generates a fixed depth map for the hands along with a mask of the hands region that will conveniently allow you to use them along with ControlNet to fix the wonky hands generated by Stable Diffusion
+
+**Node Link:** https://github.com/blessedcoolant/invoke_meshgraphormer
+
+**View**
+<img src="https://raw.githubusercontent.com/blessedcoolant/invoke_meshgraphormer/main/assets/preview.jpg" />
 
 --------------------------------
+
 ### Image and Mask Composition Pack
 
 **Description:** This is a pack of nodes for composing masks and images, including a simple text mask creator and both image and latent offset nodes. The offsets wrap around, so these can be used in conjunction with the Seamless node to progressively generate centered on different parts of the seamless tiling.
@@ -307,6 +341,20 @@ See full docs here: https://github.com/skunkworxdark/Prompt-tools-nodes/edit/mai
 
 <img src="https://github.com/skunkworxdark/match_histogram/assets/21961335/ed12f329-a0ef-444a-9bae-129ed60d6097" width="300" />
 
+--------------------------------
+### Metadata Linked Nodes
+
+**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.
+
+**Node Link:** https://github.com/skunkworxdark/metadata-linked-nodes
+
 --------------------------------
 ### Negative Image
 
@@ -317,6 +365,13 @@ Node Link: https://github.com/VeyDlin/negative-image-node
 View:
 </br><img src="https://raw.githubusercontent.com/VeyDlin/negative-image-node/master/.readme/node.png" width="500" />
 
+--------------------------------
+### Nightmare Promptgen
+
+**Description:** Nightmare Prompt Generator - Uses a local text generation model to create unique imaginative (but usually nightmarish) prompts for InvokeAI. By default, it allows you to choose from some gpt-neo models I finetuned on over 2500 of my own InvokeAI prompts in Compel format, but you're able to add your own, as well. Offers support for replacing any troublesome words with a random choice from list you can also define.
+
+**Node Link:** [https://github.com/gogurtenjoyer/nightmare-promptgen](https://github.com/gogurtenjoyer/nightmare-promptgen)
+
 --------------------------------
 ### Oobabooga
 
@@ -380,6 +435,17 @@ See full docs here: https://github.com/skunkworxdark/Prompt-tools-nodes/edit/mai
 
 **Node Link:** https://github.com/fieldOfView/InvokeAI-remote_image
 
+--------------------------------
+
+### BriaAI Remove Background
+
+**Description**: Implements one click background removal with BriaAI's new version 1.4 model which seems to be be producing better results than any other previous background removal tool.
+
+**Node Link:** https://github.com/blessedcoolant/invoke_bria_rmbg
+
+**View**
+<img src="https://raw.githubusercontent.com/blessedcoolant/invoke_bria_rmbg/main/assets/preview.jpg" />
+
 --------------------------------
 ### Remove Background
 

docs/nodes/defaultNodes.md

@@ -36,6 +36,7 @@ their descriptions.
 | Integer Math                                                  | Perform basic math operations on two integers                                                                                                        |
 | Convert Image Mode                                            | Converts an image to a different mode.                                                                                                               |
 | Crop Image                                                    | Crops an image to a specified box. The box can be outside of the image.                                                                              |
+| Ideal Size                                                    | Calculates an ideal image size for latents for a first pass of a multi-pass upscaling to avoid duplication and other artifacts                       |
 | Image Hue Adjustment                                          | Adjusts the Hue of an image.                                                                                                                         |
 | Inverse Lerp Image                                            | Inverse linear interpolation of all pixels of an image                                                                                               |
 | Image Primitive                                               | An image primitive value                                                                                                                             |

docs/nodes/exampleWorkflows.md

@@ -1,6 +1,6 @@
 # Example Workflows
 
-We've curated some example workflows for you to get started with Workflows in InvokeAI
+We've curated some example workflows for you to get started with Workflows in InvokeAI! These can also be found in the Workflow Library, located in the Workflow Editor of Invoke.
 
 To use them, right click on your desired workflow, follow the link to GitHub and click the "⬇" button to download the raw file. You can then use the "Load Workflow" functionality in InvokeAI to load the workflow and start generating images!
 

docs/other/CONTRIBUTORS.md

@@ -13,46 +13,69 @@ We thank them for all of their time and hard work.
 
 - [Lincoln D. Stein](mailto:lincoln.stein@gmail.com)
 
-## **Current core team**
+## **Current Core Team**
 
 * @lstein (Lincoln Stein) - Co-maintainer
 * @blessedcoolant - Co-maintainer
 * @hipsterusername (Kent Keirsey) - Co-maintainer, CEO, Positive Vibes
 * @psychedelicious (Spencer Mabrito) - Web Team Leader
-* @Kyle0654 (Kyle Schouviller) - Node Architect and General Backend Wizard
-* @damian0815 - Attention Systems and Compel Maintainer
-* @ebr (Eugene Brodsky) - Cloud/DevOps/Sofware engineer; your friendly neighbourhood cluster-autoscaler
-* @genomancer (Gregg Helt) - Controlnet support
-* @StAlKeR7779 (Sergey Borisov) - Torch stack, ONNX, model management, optimization
+* @chainchompa (Jennifer Player) - Web Development & Chain-Chomping
+* @josh is toast (Josh Corbett) - Web Development
 * @cheerio (Mary Rogers) - Lead Engineer & Web App Development
+* @ebr (Eugene Brodsky) - Cloud/DevOps/Sofware engineer; your friendly neighbourhood cluster-autoscaler
+* @sunija - Standalone version
+* @genomancer (Gregg Helt) - Controlnet support
 * @brandon (Brandon Rising) - Platform, Infrastructure, Backend Systems
 * @ryanjdick (Ryan Dick) - Machine Learning & Training
-* @millu (Millun Atluri) - Community Manager, Documentation, Node-wrangler
-* @chainchompa (Jennifer Player) - Web Development & Chain-Chomping
+* @JPPhoto - Core image generation nodes
+* @dunkeroni - Image generation backend
+* @SkunkWorxDark - Image generation backend
 * @keturn (Kevin Turner) - Diffusers
+* @millu (Millun Atluri) - Community Wizard, Documentation, Node-wrangler, 
+* @glimmerleaf (Devon Hopkins) - Community Wizard
 * @gogurt enjoyer - Discord moderator and end user support
 * @whosawhatsis - Discord moderator and end user support
 * @dwinrger - Discord moderator and end user support
 * @526christian - Discord moderator and end user support
+* @harvester62 -  Discord moderator and end user support
+
+
+## **Honored Team Alumni**
+
+* @StAlKeR7779 (Sergey Borisov) - Torch stack, ONNX, model management, optimization
+* @damian0815 - Attention Systems and Compel Maintainer
+* @netsvetaev (Artur) - Localization support
+* @Kyle0654 (Kyle Schouviller) - Node Architect and General Backend Wizard
+* @tildebyte - Installation and configuration
+* @mauwii (Matthias Wilde) - Installation, release, continuous integration
+
 
 ## **Full List of Contributors by Commit Name**
 
+- 이승석
 - AbdBarho
 - ablattmann
 - AdamOStark
 - Adam Rice
 - Airton Silva
+- Aldo Hoeben
 - Alexander Eichhorn
 - Alexandre D. Roberge
+- Alexandre Macabies
+- Alfie John
 - Andreas Rozek
 - Andre LaBranche
 - Andy Bearman
 - Andy Luhrs
 - Andy Pilate
+- Anonymous
+- Anthony Monthe
 - Any-Winter-4079
 - apolinario
+- Ar7ific1al
 - ArDiouscuros
 - Armando C. Santisbon
+- Arnold Cordewiner
 - Arthur Holstvoogd
 - artmen1516
 - Artur
@@ -64,13 +87,16 @@ We thank them for all of their time and hard work.
 - blhook
 - BlueAmulet
 - Bouncyknighter
+- Brandon
 - Brandon Rising
 - Brent Ozar
 - Brian Racer
 - bsilvereagle
 - c67e708d
+- camenduru
 - CapableWeb
 - Carson Katri
+- chainchompa
 - Chloe
 - Chris Dawson
 - Chris Hayes
@@ -86,30 +112,45 @@ We thank them for all of their time and hard work.
 - cpacker
 - Cragin Godley
 - creachec
+- CrypticWit
+- d8ahazard
+- damian
+- damian0815
+- Damian at mba
 - Damian Stewart
 - Daniel Manzke
 - Danny Beer
 - Dan Sully
+- Darren Ringer
 - David Burnett
 - David Ford
 - David Regla
+- David Sisco
 - David Wager
 - Daya Adianto
 - db3000
+- DekitaRPG
 - Denis Olshin
 - Dennis
+- dependabot[bot]
+- Dmitry Parnas
+- Dobrynia100
 - Dominic Letz
 - DrGunnarMallon
+- Drun555
+- dunkeroni
 - Edward Johan
 - elliotsayes
 - Elrik
 - ElrikUnderlake
 - Eric Khun
 - Eric Wolf
+- Eugene
 - Eugene Brodsky
 - ExperimentalCyborg
 - Fabian Bahl
 - Fabio 'MrWHO' Torchetti
+- Fattire
 - fattire
 - Felipe Nogueira
 - Félix Sanz
@@ -118,8 +159,12 @@ We thank them for all of their time and hard work.
 - gabrielrotbart
 - gallegonovato
 - Gérald LONLAS
+- Gille
 - GitHub Actions Bot
+- glibesyck
 - gogurtenjoyer
+- Gohsuke Shimada
+- greatwolf
 - greentext2
 - Gregg Helt
 - H4rk
@@ -131,6 +176,7 @@ We thank them for all of their time and hard work.
 - Hosted Weblate
 - Iman Karim
 - ismail ihsan bülbül
+- ItzAttila
 - Ivan Efimov
 - jakehl
 - Jakub Kolčář
@@ -141,6 +187,7 @@ We thank them for all of their time and hard work.
 - Jason Toffaletti
 - Jaulustus
 - Jeff Mahoney
+- Jennifer Player
 - jeremy
 - Jeremy Clark
 - JigenD
@@ -148,19 +195,26 @@ We thank them for all of their time and hard work.
 - Johan Roxendal
 - Johnathon Selstad
 - Jonathan
+- Jordan Hewitt
 - Joseph Dries III
+- Josh Corbett
 - JPPhoto
 - jspraul
+- junzi
 - Justin Wong
 - Juuso V
 - Kaspar Emanuel
 - Katsuyuki-Karasawa
+- Keerigan45
 - Kent Keirsey
+- Kevin Brack
 - Kevin Coakley
 - Kevin Gibbons
 - Kevin Schaul
 - Kevin Turner
+- Kieran Klaassen
 - krummrey
+- Kyle
 - Kyle Lacy
 - Kyle Schouviller
 - Lawrence Norton
@@ -171,10 +225,15 @@ We thank them for all of their time and hard work.
 - Lynne Whitehorn
 - majick
 - Marco Labarile
+- Marta Nahorniuk
 - Martin Kristiansen
+- Mary Hipp
+- maryhipp
 - Mary Hipp Rogers
+- mastercaster
 - mastercaster9000
 - Matthias Wild
+- mauwii
 - michaelk71
 - mickr777
 - Mihai
@@ -182,11 +241,15 @@ We thank them for all of their time and hard work.
 - Mikhail Tishin
 - Millun Atluri
 - Minjune Song
+- Mitchell Allain
 - mitien
 - mofuzz
 - Muhammad Usama
 - Name
 - _nderscore
+- Neil Wang
+- nekowaiz
+- nemuruibai
 - Netzer R
 - Nicholas Koh
 - Nicholas Körfer
@@ -197,9 +260,11 @@ We thank them for all of their time and hard work.
 - ofirkris
 - Olivier Louvignes
 - owenvincent
+- pand4z31
 - Patrick Esser
 - Patrick Tien
 - Patrick von Platen
+- Paul Curry
 - Paul Sajna
 - pejotr
 - Peter Baylies
@@ -207,6 +272,7 @@ We thank them for all of their time and hard work.
 - plucked
 - prixt
 - psychedelicious
+- psychedelicious@windows
 - Rainer Bernhardt
 - Riccardo Giovanetti
 - Rich Jones
@@ -215,16 +281,22 @@ We thank them for all of their time and hard work.
 - Robert Bolender
 - Robin Rombach
 - Rohan Barar
+- Rohinish
 - rpagliuca
 - rromb
 - Rupesh Sreeraman
+- Ryan
 - Ryan Cao
+- Ryan Dick
 - Saifeddine
 - Saifeddine ALOUI
+- Sam
 - SammCheese
+- Sam McLeod
 - Sammy
 - sammyf
 - Samuel Husso
+- Saurav Maheshkar
 - Scott Lahteine
 - Sean McLellan
 - Sebastian Aigner
@@ -232,16 +304,21 @@ We thank them for all of their time and hard work.
 - Sergey Krashevich
 - Shapor Naghibzadeh
 - Shawn Zhong
+- Simona Liliac
 - Simon Vans-Colina
 - skunkworxdark
 - slashtechno
+- SoheilRezaei
+- Song, Pengcheng
 - spezialspezial
 - ssantos
 - StAlKeR7779
+- Stefan Tobler
 - Stephan Koglin-Fischer
 - SteveCaruso
 - Steve Martinelli
 - Steven Frank
+- Surisen
 - System X - Files
 - Taylor Kems
 - techicode
@@ -260,26 +337,34 @@ We thank them for all of their time and hard work.
 - tyler
 - unknown
 - user1
+- vedant-3010
 - Vedant Madane
 - veprogames
 - wa.code
 - wfng92
+- whjms
 - whosawhatsis
 - Will
 - William Becher
 - William Chong
+- Wilson E. Alvarez
+- woweenie
+- Wubbbi
 - xra
 - Yeung Yiu Hung
 - ymgenesis
 - Yorzaren
 - Yosuke Shinya
 - yun saki
+- ZachNagengast
 - Zadagu
 - zeptofine
+- Zerdoumi
+- Васянатор
 - 冯不游
 - 唐澤 克幸
 
-## **Original CompVis Authors**
+## **Original CompVis (Stable Diffusion) Authors**
 
 - [Robin Rombach](https://github.com/rromb)
 - [Patrick von Platen](https://github.com/patrickvonplaten)

docs/stylesheets/extra.css

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

docs/workflows/Text_to_Image.json

@@ -1,8 +1,8 @@
 {
-  "name": "Text to Image",
+  "name": "Text to Image - SD1.5",
   "author": "InvokeAI",
   "description": "Sample text to image workflow for Stable Diffusion 1.5/2",
-  "version": "1.0.1",
+  "version": "1.1.0",
   "contact": "invoke@invoke.ai",
   "tags": "text2image, SD1.5, SD2, default",
   "notes": "",
@@ -18,10 +18,19 @@
     {
       "nodeId": "93dc02a4-d05b-48ed-b99c-c9b616af3402",
       "fieldName": "prompt"
+    },
+    {
+      "nodeId": "55705012-79b9-4aac-9f26-c0b10309785b",
+      "fieldName": "width"
+    },
+    {
+      "nodeId": "55705012-79b9-4aac-9f26-c0b10309785b",
+      "fieldName": "height"
     }
   ],
   "meta": {
-    "version": "1.0.0"
+    "category": "default",
+    "version": "2.0.0"
   },
   "nodes": [
     {
@@ -30,44 +39,56 @@
       "data": {
         "id": "93dc02a4-d05b-48ed-b99c-c9b616af3402",
         "type": "compel",
+        "label": "Negative Compel Prompt",
+        "isOpen": true,
+        "notes": "",
+        "isIntermediate": true,
+        "useCache": true,
+        "version": "1.0.0",
+        "nodePack": "invokeai",
         "inputs": {
           "prompt": {
             "id": "7739aff6-26cb-4016-8897-5a1fb2305e4e",
             "name": "prompt",
-            "type": "string",
             "fieldKind": "input",
             "label": "Negative Prompt",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "StringField"
+            },
             "value": ""
           },
           "clip": {
             "id": "48d23dce-a6ae-472a-9f8c-22a714ea5ce0",
             "name": "clip",
-            "type": "ClipField",
             "fieldKind": "input",
-            "label": ""
+            "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "ClipField"
+            }
           }
         },
         "outputs": {
           "conditioning": {
             "id": "37cf3a9d-f6b7-4b64-8ff6-2558c5ecc447",
             "name": "conditioning",
-            "type": "ConditioningField",
-            "fieldKind": "output"
+            "fieldKind": "output",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "ConditioningField"
+            }
           }
-        },
-        "label": "Negative Compel Prompt",
-        "isOpen": true,
-        "notes": "",
-        "embedWorkflow": false,
-        "isIntermediate": true,
-        "useCache": true,
-        "version": "1.0.0"
+        }
       },
       "width": 320,
-      "height": 261,
+      "height": 259,
       "position": {
-        "x": 995.7263915923627,
-        "y": 239.67783573351227
+        "x": 1000,
+        "y": 350
       }
     },
     {
@@ -76,37 +97,60 @@
       "data": {
         "id": "55705012-79b9-4aac-9f26-c0b10309785b",
         "type": "noise",
+        "label": "",
+        "isOpen": true,
+        "notes": "",
+        "isIntermediate": true,
+        "useCache": true,
+        "version": "1.0.1",
+        "nodePack": "invokeai",
         "inputs": {
           "seed": {
             "id": "6431737c-918a-425d-a3b4-5d57e2f35d4d",
             "name": "seed",
-            "type": "integer",
             "fieldKind": "input",
             "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "IntegerField"
+            },
             "value": 0
           },
           "width": {
             "id": "38fc5b66-fe6e-47c8-bba9-daf58e454ed7",
             "name": "width",
-            "type": "integer",
             "fieldKind": "input",
             "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "IntegerField"
+            },
             "value": 512
           },
           "height": {
             "id": "16298330-e2bf-4872-a514-d6923df53cbb",
             "name": "height",
-            "type": "integer",
             "fieldKind": "input",
             "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "IntegerField"
+            },
             "value": 512
           },
           "use_cpu": {
             "id": "c7c436d3-7a7a-4e76-91e4-c6deb271623c",
             "name": "use_cpu",
-            "type": "boolean",
             "fieldKind": "input",
             "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "BooleanField"
+            },
             "value": true
           }
         },
@@ -114,35 +158,40 @@
           "noise": {
             "id": "50f650dc-0184-4e23-a927-0497a96fe954",
             "name": "noise",
-            "type": "LatentsField",
-            "fieldKind": "output"
+            "fieldKind": "output",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "LatentsField"
+            }
           },
           "width": {
             "id": "bb8a452b-133d-42d1-ae4a-3843d7e4109a",
             "name": "width",
-            "type": "integer",
-            "fieldKind": "output"
+            "fieldKind": "output",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "IntegerField"
+            }
           },
           "height": {
             "id": "35cfaa12-3b8b-4b7a-a884-327ff3abddd9",
             "name": "height",
-            "type": "integer",
-            "fieldKind": "output"
+            "fieldKind": "output",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "IntegerField"
+            }
           }
-        },
-        "label": "",
-        "isOpen": true,
-        "notes": "",
-        "embedWorkflow": false,
-        "isIntermediate": true,
-        "useCache": true,
-        "version": "1.0.0"
+        }
       },
       "width": 320,
-      "height": 389,
+      "height": 388,
       "position": {
-        "x": 993.4442117555518,
-        "y": 605.6757415334787
+        "x": 600,
+        "y": 325
       }
     },
     {
@@ -151,13 +200,24 @@
       "data": {
         "id": "c8d55139-f380-4695-b7f2-8b3d1e1e3db8",
         "type": "main_model_loader",
+        "label": "",
+        "isOpen": true,
+        "notes": "",
+        "isIntermediate": true,
+        "useCache": true,
+        "version": "1.0.0",
+        "nodePack": "invokeai",
         "inputs": {
           "model": {
             "id": "993eabd2-40fd-44fe-bce7-5d0c7075ddab",
             "name": "model",
-            "type": "MainModelField",
             "fieldKind": "input",
             "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "MainModelField"
+            },
             "value": {
               "model_name": "stable-diffusion-v1-5",
               "base_model": "sd-1",
@@ -169,35 +229,40 @@
           "unet": {
             "id": "5c18c9db-328d-46d0-8cb9-143391c410be",
             "name": "unet",
-            "type": "UNetField",
-            "fieldKind": "output"
+            "fieldKind": "output",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "UNetField"
+            }
           },
           "clip": {
             "id": "6effcac0-ec2f-4bf5-a49e-a2c29cf921f4",
             "name": "clip",
-            "type": "ClipField",
-            "fieldKind": "output"
+            "fieldKind": "output",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "ClipField"
+            }
           },
           "vae": {
             "id": "57683ba3-f5f5-4f58-b9a2-4b83dacad4a1",
             "name": "vae",
-            "type": "VaeField",
-            "fieldKind": "output"
+            "fieldKind": "output",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "VaeField"
+            }
           }
-        },
-        "label": "",
-        "isOpen": true,
-        "notes": "",
-        "embedWorkflow": false,
-        "isIntermediate": true,
-        "useCache": true,
-        "version": "1.0.0"
+        }
       },
       "width": 320,
       "height": 226,
       "position": {
-        "x": 163.04436745878343,
-        "y": 254.63156870373479
+        "x": 600,
+        "y": 25
       }
     },
     {
@@ -206,44 +271,56 @@
       "data": {
         "id": "7d8bf987-284f-413a-b2fd-d825445a5d6c",
         "type": "compel",
+        "label": "Positive Compel Prompt",
+        "isOpen": true,
+        "notes": "",
+        "isIntermediate": true,
+        "useCache": true,
+        "version": "1.0.0",
+        "nodePack": "invokeai",
         "inputs": {
           "prompt": {
             "id": "7739aff6-26cb-4016-8897-5a1fb2305e4e",
             "name": "prompt",
-            "type": "string",
             "fieldKind": "input",
             "label": "Positive Prompt",
-            "value": ""
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "StringField"
+            },
+            "value": "Super cute tiger cub, national geographic award-winning photograph"
           },
           "clip": {
             "id": "48d23dce-a6ae-472a-9f8c-22a714ea5ce0",
             "name": "clip",
-            "type": "ClipField",
             "fieldKind": "input",
-            "label": ""
+            "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "ClipField"
+            }
           }
         },
         "outputs": {
           "conditioning": {
             "id": "37cf3a9d-f6b7-4b64-8ff6-2558c5ecc447",
             "name": "conditioning",
-            "type": "ConditioningField",
-            "fieldKind": "output"
+            "fieldKind": "output",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "ConditioningField"
+            }
           }
-        },
-        "label": "Positive Compel Prompt",
-        "isOpen": true,
-        "notes": "",
-        "embedWorkflow": false,
-        "isIntermediate": true,
-        "useCache": true,
-        "version": "1.0.0"
+        }
       },
       "width": 320,
-      "height": 261,
+      "height": 259,
       "position": {
-        "x": 595.7263915923627,
-        "y": 239.67783573351227
+        "x": 1000,
+        "y": 25
       }
     },
     {
@@ -252,21 +329,36 @@
       "data": {
         "id": "ea94bc37-d995-4a83-aa99-4af42479f2f2",
         "type": "rand_int",
+        "label": "Random Seed",
+        "isOpen": false,
+        "notes": "",
+        "isIntermediate": true,
+        "useCache": false,
+        "version": "1.0.0",
+        "nodePack": "invokeai",
         "inputs": {
           "low": {
             "id": "3ec65a37-60ba-4b6c-a0b2-553dd7a84b84",
             "name": "low",
-            "type": "integer",
             "fieldKind": "input",
             "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "IntegerField"
+            },
             "value": 0
           },
           "high": {
             "id": "085f853a-1a5f-494d-8bec-e4ba29a3f2d1",
             "name": "high",
-            "type": "integer",
             "fieldKind": "input",
             "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "IntegerField"
+            },
             "value": 2147483647
           }
         },
@@ -274,23 +366,20 @@
           "value": {
             "id": "812ade4d-7699-4261-b9fc-a6c9d2ab55ee",
             "name": "value",
-            "type": "integer",
-            "fieldKind": "output"
+            "fieldKind": "output",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "IntegerField"
+            }
           }
-        },
-        "label": "Random Seed",
-        "isOpen": true,
-        "notes": "",
-        "embedWorkflow": false,
-        "isIntermediate": true,
-        "useCache": false,
-        "version": "1.0.0"
+        }
       },
       "width": 320,
-      "height": 218,
+      "height": 32,
       "position": {
-        "x": 541.094822888628,
-        "y": 694.5704476446829
+        "x": 600,
+        "y": 275
       }
     },
     {
@@ -299,144 +388,224 @@
       "data": {
         "id": "eea2702a-19fb-45b5-9d75-56b4211ec03c",
         "type": "denoise_latents",
+        "label": "",
+        "isOpen": true,
+        "notes": "",
+        "isIntermediate": true,
+        "useCache": true,
+        "version": "1.5.0",
+        "nodePack": "invokeai",
         "inputs": {
           "positive_conditioning": {
             "id": "90b7f4f8-ada7-4028-8100-d2e54f192052",
             "name": "positive_conditioning",
-            "type": "ConditioningField",
             "fieldKind": "input",
-            "label": ""
+            "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "ConditioningField"
+            }
           },
           "negative_conditioning": {
             "id": "9393779e-796c-4f64-b740-902a1177bf53",
             "name": "negative_conditioning",
-            "type": "ConditioningField",
             "fieldKind": "input",
-            "label": ""
+            "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "ConditioningField"
+            }
           },
           "noise": {
             "id": "8e17f1e5-4f98-40b1-b7f4-86aeeb4554c1",
             "name": "noise",
-            "type": "LatentsField",
             "fieldKind": "input",
-            "label": ""
+            "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "LatentsField"
+            }
           },
           "steps": {
             "id": "9b63302d-6bd2-42c9-ac13-9b1afb51af88",
             "name": "steps",
-            "type": "integer",
             "fieldKind": "input",
             "label": "",
-            "value": 10
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "IntegerField"
+            },
+            "value": 50
           },
           "cfg_scale": {
             "id": "87dd04d3-870e-49e1-98bf-af003a810109",
             "name": "cfg_scale",
-            "type": "FloatPolymorphic",
             "fieldKind": "input",
             "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": true,
+              "name": "FloatField"
+            },
             "value": 7.5
           },
           "denoising_start": {
             "id": "f369d80f-4931-4740-9bcd-9f0620719fab",
             "name": "denoising_start",
-            "type": "float",
             "fieldKind": "input",
             "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "FloatField"
+            },
             "value": 0
           },
           "denoising_end": {
             "id": "747d10e5-6f02-445c-994c-0604d814de8c",
             "name": "denoising_end",
-            "type": "float",
             "fieldKind": "input",
             "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "FloatField"
+            },
             "value": 1
           },
           "scheduler": {
             "id": "1de84a4e-3a24-4ec8-862b-16ce49633b9b",
             "name": "scheduler",
-            "type": "Scheduler",
             "fieldKind": "input",
             "label": "",
-            "value": "euler"
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "SchedulerField"
+            },
+            "value": "unipc"
           },
           "unet": {
             "id": "ffa6fef4-3ce2-4bdb-9296-9a834849489b",
             "name": "unet",
-            "type": "UNetField",
             "fieldKind": "input",
-            "label": ""
+            "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "UNetField"
+            }
           },
           "control": {
             "id": "077b64cb-34be-4fcc-83f2-e399807a02bd",
             "name": "control",
-            "type": "ControlPolymorphic",
             "fieldKind": "input",
-            "label": ""
+            "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": true,
+              "name": "ControlField"
+            }
           },
           "ip_adapter": {
             "id": "1d6948f7-3a65-4a65-a20c-768b287251aa",
             "name": "ip_adapter",
-            "type": "IPAdapterPolymorphic",
             "fieldKind": "input",
-            "label": ""
+            "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": true,
+              "name": "IPAdapterField"
+            }
           },
           "t2i_adapter": {
             "id": "75e67b09-952f-4083-aaf4-6b804d690412",
             "name": "t2i_adapter",
-            "type": "T2IAdapterPolymorphic",
             "fieldKind": "input",
-            "label": ""
+            "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": true,
+              "name": "T2IAdapterField"
+            }
+          },
+          "cfg_rescale_multiplier": {
+            "id": "9101f0a6-5fe0-4826-b7b3-47e5d506826c",
+            "name": "cfg_rescale_multiplier",
+            "fieldKind": "input",
+            "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "FloatField"
+            },
+            "value": 0
           },
           "latents": {
             "id": "334d4ba3-5a99-4195-82c5-86fb3f4f7d43",
             "name": "latents",
-            "type": "LatentsField",
             "fieldKind": "input",
-            "label": ""
+            "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "LatentsField"
+            }
           },
           "denoise_mask": {
             "id": "0d3dbdbf-b014-4e95-8b18-ff2ff9cb0bfa",
             "name": "denoise_mask",
-            "type": "DenoiseMaskField",
             "fieldKind": "input",
-            "label": ""
+            "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "DenoiseMaskField"
+            }
           }
         },
         "outputs": {
           "latents": {
             "id": "70fa5bbc-0c38-41bb-861a-74d6d78d2f38",
             "name": "latents",
-            "type": "LatentsField",
-            "fieldKind": "output"
+            "fieldKind": "output",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "LatentsField"
+            }
           },
           "width": {
             "id": "98ee0e6c-82aa-4e8f-8be5-dc5f00ee47f0",
             "name": "width",
-            "type": "integer",
-            "fieldKind": "output"
+            "fieldKind": "output",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "IntegerField"
+            }
           },
           "height": {
             "id": "e8cb184a-5e1a-47c8-9695-4b8979564f5d",
             "name": "height",
-            "type": "integer",
-            "fieldKind": "output"
+            "fieldKind": "output",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "IntegerField"
+            }
           }
-        },
-        "label": "",
-        "isOpen": true,
-        "notes": "",
-        "embedWorkflow": false,
-        "isIntermediate": true,
-        "useCache": true,
-        "version": "1.4.0"
+        }
       },
       "width": 320,
-      "height": 646,
+      "height": 703,
       "position": {
-        "x": 1476.5794704734735,
-        "y": 256.80174342731783
+        "x": 1400,
+        "y": 25
       }
     },
     {
@@ -445,153 +614,185 @@
       "data": {
         "id": "58c957f5-0d01-41fc-a803-b2bbf0413d4f",
         "type": "l2i",
+        "label": "",
+        "isOpen": true,
+        "notes": "",
+        "isIntermediate": false,
+        "useCache": true,
+        "version": "1.2.0",
+        "nodePack": "invokeai",
         "inputs": {
           "metadata": {
             "id": "ab375f12-0042-4410-9182-29e30db82c85",
             "name": "metadata",
-            "type": "MetadataField",
             "fieldKind": "input",
-            "label": ""
+            "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "MetadataField"
+            }
           },
           "latents": {
             "id": "3a7e7efd-bff5-47d7-9d48-615127afee78",
             "name": "latents",
-            "type": "LatentsField",
             "fieldKind": "input",
-            "label": ""
+            "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "LatentsField"
+            }
           },
           "vae": {
             "id": "a1f5f7a1-0795-4d58-b036-7820c0b0ef2b",
             "name": "vae",
-            "type": "VaeField",
             "fieldKind": "input",
-            "label": ""
+            "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "VaeField"
+            }
           },
           "tiled": {
             "id": "da52059a-0cee-4668-942f-519aa794d739",
             "name": "tiled",
-            "type": "boolean",
             "fieldKind": "input",
             "label": "",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "BooleanField"
+            },
             "value": false
           },
           "fp32": {
             "id": "c4841df3-b24e-4140-be3b-ccd454c2522c",
             "name": "fp32",
-            "type": "boolean",
             "fieldKind": "input",
             "label": "",
-            "value": false
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "BooleanField"
+            },
+            "value": true
           }
         },
         "outputs": {
           "image": {
             "id": "72d667d0-cf85-459d-abf2-28bd8b823fe7",
             "name": "image",
-            "type": "ImageField",
-            "fieldKind": "output"
+            "fieldKind": "output",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "ImageField"
+            }
           },
           "width": {
             "id": "c8c907d8-1066-49d1-b9a6-83bdcd53addc",
             "name": "width",
-            "type": "integer",
-            "fieldKind": "output"
+            "fieldKind": "output",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "IntegerField"
+            }
           },
           "height": {
             "id": "230f359c-b4ea-436c-b372-332d7dcdca85",
             "name": "height",
-            "type": "integer",
-            "fieldKind": "output"
+            "fieldKind": "output",
+            "type": {
+              "isCollection": false,
+              "isCollectionOrScalar": false,
+              "name": "IntegerField"
+            }
           }
-        },
-        "label": "",
-        "isOpen": true,
-        "notes": "",
-        "embedWorkflow": false,
-        "isIntermediate": false,
-        "useCache": true,
-        "version": "1.0.0"
+        }
       },
       "width": 320,
-      "height": 267,
+      "height": 266,
       "position": {
-        "x": 2037.9648469717395,
-        "y": 426.10844427600136
+        "x": 1800,
+        "y": 25
       }
     }
   ],
   "edges": [
     {
-      "source": "ea94bc37-d995-4a83-aa99-4af42479f2f2",
-      "sourceHandle": "value",
-      "target": "55705012-79b9-4aac-9f26-c0b10309785b",
-      "targetHandle": "seed",
       "id": "reactflow__edge-ea94bc37-d995-4a83-aa99-4af42479f2f2value-55705012-79b9-4aac-9f26-c0b10309785bseed",
-      "type": "default"
+      "source": "ea94bc37-d995-4a83-aa99-4af42479f2f2",
+      "target": "55705012-79b9-4aac-9f26-c0b10309785b",
+      "type": "default",
+      "sourceHandle": "value",
+      "targetHandle": "seed"
     },
     {
-      "source": "c8d55139-f380-4695-b7f2-8b3d1e1e3db8",
-      "sourceHandle": "clip",
-      "target": "7d8bf987-284f-413a-b2fd-d825445a5d6c",
-      "targetHandle": "clip",
       "id": "reactflow__edge-c8d55139-f380-4695-b7f2-8b3d1e1e3db8clip-7d8bf987-284f-413a-b2fd-d825445a5d6cclip",
-      "type": "default"
-    },
-    {
       "source": "c8d55139-f380-4695-b7f2-8b3d1e1e3db8",
+      "target": "7d8bf987-284f-413a-b2fd-d825445a5d6c",
+      "type": "default",
       "sourceHandle": "clip",
-      "target": "93dc02a4-d05b-48ed-b99c-c9b616af3402",
-      "targetHandle": "clip",
+      "targetHandle": "clip"
+    },
+    {
       "id": "reactflow__edge-c8d55139-f380-4695-b7f2-8b3d1e1e3db8clip-93dc02a4-d05b-48ed-b99c-c9b616af3402clip",
-      "type": "default"
+      "source": "c8d55139-f380-4695-b7f2-8b3d1e1e3db8",
+      "target": "93dc02a4-d05b-48ed-b99c-c9b616af3402",
+      "type": "default",
+      "sourceHandle": "clip",
+      "targetHandle": "clip"
     },
     {
-      "source": "55705012-79b9-4aac-9f26-c0b10309785b",
-      "sourceHandle": "noise",
-      "target": "eea2702a-19fb-45b5-9d75-56b4211ec03c",
-      "targetHandle": "noise",
       "id": "reactflow__edge-55705012-79b9-4aac-9f26-c0b10309785bnoise-eea2702a-19fb-45b5-9d75-56b4211ec03cnoise",
-      "type": "default"
+      "source": "55705012-79b9-4aac-9f26-c0b10309785b",
+      "target": "eea2702a-19fb-45b5-9d75-56b4211ec03c",
+      "type": "default",
+      "sourceHandle": "noise",
+      "targetHandle": "noise"
     },
     {
-      "source": "7d8bf987-284f-413a-b2fd-d825445a5d6c",
-      "sourceHandle": "conditioning",
-      "target": "eea2702a-19fb-45b5-9d75-56b4211ec03c",
-      "targetHandle": "positive_conditioning",
       "id": "reactflow__edge-7d8bf987-284f-413a-b2fd-d825445a5d6cconditioning-eea2702a-19fb-45b5-9d75-56b4211ec03cpositive_conditioning",
-      "type": "default"
-    },
-    {
-      "source": "93dc02a4-d05b-48ed-b99c-c9b616af3402",
+      "source": "7d8bf987-284f-413a-b2fd-d825445a5d6c",
+      "target": "eea2702a-19fb-45b5-9d75-56b4211ec03c",
+      "type": "default",
       "sourceHandle": "conditioning",
-      "target": "eea2702a-19fb-45b5-9d75-56b4211ec03c",
-      "targetHandle": "negative_conditioning",
+      "targetHandle": "positive_conditioning"
+    },
+    {
       "id": "reactflow__edge-93dc02a4-d05b-48ed-b99c-c9b616af3402conditioning-eea2702a-19fb-45b5-9d75-56b4211ec03cnegative_conditioning",
-      "type": "default"
-    },
-    {
-      "source": "c8d55139-f380-4695-b7f2-8b3d1e1e3db8",
-      "sourceHandle": "unet",
+      "source": "93dc02a4-d05b-48ed-b99c-c9b616af3402",
       "target": "eea2702a-19fb-45b5-9d75-56b4211ec03c",
-      "targetHandle": "unet",
+      "type": "default",
+      "sourceHandle": "conditioning",
+      "targetHandle": "negative_conditioning"
+    },
+    {
       "id": "reactflow__edge-c8d55139-f380-4695-b7f2-8b3d1e1e3db8unet-eea2702a-19fb-45b5-9d75-56b4211ec03cunet",
-      "type": "default"
-    },
-    {
-      "source": "eea2702a-19fb-45b5-9d75-56b4211ec03c",
-      "sourceHandle": "latents",
-      "target": "58c957f5-0d01-41fc-a803-b2bbf0413d4f",
-      "targetHandle": "latents",
-      "id": "reactflow__edge-eea2702a-19fb-45b5-9d75-56b4211ec03clatents-58c957f5-0d01-41fc-a803-b2bbf0413d4flatents",
-      "type": "default"
-    },
-    {
       "source": "c8d55139-f380-4695-b7f2-8b3d1e1e3db8",
-      "sourceHandle": "vae",
+      "target": "eea2702a-19fb-45b5-9d75-56b4211ec03c",
+      "type": "default",
+      "sourceHandle": "unet",
+      "targetHandle": "unet"
+    },
+    {
+      "id": "reactflow__edge-eea2702a-19fb-45b5-9d75-56b4211ec03clatents-58c957f5-0d01-41fc-a803-b2bbf0413d4flatents",
+      "source": "eea2702a-19fb-45b5-9d75-56b4211ec03c",
       "target": "58c957f5-0d01-41fc-a803-b2bbf0413d4f",
-      "targetHandle": "vae",
+      "type": "default",
+      "sourceHandle": "latents",
+      "targetHandle": "latents"
+    },
+    {
       "id": "reactflow__edge-c8d55139-f380-4695-b7f2-8b3d1e1e3db8vae-58c957f5-0d01-41fc-a803-b2bbf0413d4fvae",
-      "type": "default"
+      "source": "c8d55139-f380-4695-b7f2-8b3d1e1e3db8",
+      "target": "58c957f5-0d01-41fc-a803-b2bbf0413d4f",
+      "type": "default",
+      "sourceHandle": "vae",
+      "targetHandle": "vae"
     }
   ]
-}
+}

installer/create_installer.sh

@@ -14,11 +14,19 @@ function is_bin_in_path {
 }
 
 function git_show {
-    git show -s --format='%h %s' $1
+    git show -s --format=oneline --abbrev-commit "$1" | cat
 }
 
+if [[ -v "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}"
+    exit -1
+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..."
@@ -32,13 +40,6 @@ if ! is_bin_in_path python && is_bin_in_path python3; then
     }
 fi
 
-if [[ -v "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}"
-    exit -1
-fi
-
 VERSION=$(
     cd ..
     python -c "from invokeai.version import __version__ as version; print(version)"
@@ -47,38 +48,9 @@ PATCH=""
 VERSION="v${VERSION}${PATCH}"
 
 echo -e "${BGREEN}HEAD${RESET}:"
-git_show
+git_show HEAD
 echo
 
-# ---------------------- FRONTEND ----------------------
-
-pushd ../invokeai/frontend/web >/dev/null
-echo
-echo "Installing frontend dependencies..."
-echo
-pnpm i --frozen-lockfile
-echo
-echo "Building frontend..."
-echo
-pnpm build
-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 [[ $(python -c 'from importlib.util import find_spec; print(find_spec("build") is None)') == "True" ]]; then
-    pip install --user build
-fi
-
-rm -rf ../build
-
-python -m build --wheel --outdir dist/ ../.
-
 # ----------------------
 
 echo
@@ -91,12 +63,11 @@ rm -rf InvokeAI-Installer
 
 # copy content
 mkdir InvokeAI-Installer
-for f in templates lib *.txt *.reg; do
+for f in templates *.txt *.reg; do
     cp -r ${f} InvokeAI-Installer/
 done
-
-# Move the wheel
-mv dist/*.whl InvokeAI-Installer/lib/
+mkdir InvokeAI-Installer/lib
+cp lib/*.py InvokeAI-Installer/lib
 
 # Install scripts
 # Mac/Linux
@@ -104,13 +75,13 @@ cp install.sh.in InvokeAI-Installer/install.sh
 chmod a+x InvokeAI-Installer/install.sh
 
 # Windows
-perl -p -e "s/^set INVOKEAI_VERSION=.*/set INVOKEAI_VERSION=$VERSION/" install.bat.in >InvokeAI-Installer/install.bat
+cp install.bat.in InvokeAI-Installer/install.bat
 cp WinLongPathsEnabled.reg InvokeAI-Installer/
 
 # Zip everything up
 zip -r InvokeAI-installer-$VERSION.zip InvokeAI-Installer
 
 # clean up
-rm -rf InvokeAI-Installer tmp dist
+rm -rf InvokeAI-Installer tmp dist ../invokeai/frontend/web/dist/
 
 exit 0

installer/install.bat.in

@@ -15,7 +15,6 @@ if "%1" == "use-cache" (
 @rem Config
 @rem The version in the next line is replaced by an up to date release number
 @rem when create_installer.sh is run. Change the release number there.
-set INVOKEAI_VERSION=latest
 set INSTRUCTIONS=https://invoke-ai.github.io/InvokeAI/installation/INSTALL_AUTOMATED/
 set TROUBLESHOOTING=https://invoke-ai.github.io/InvokeAI/installation/INSTALL_AUTOMATED/#troubleshooting
 set PYTHON_URL=https://www.python.org/downloads/windows/

installer/lib/installer.py

@@ -11,7 +11,7 @@ import sys
 import venv
 from pathlib import Path
 from tempfile import TemporaryDirectory
-from typing import Union
+from typing import Optional, Tuple
 
 SUPPORTED_PYTHON = ">=3.10.0,<=3.11.100"
 INSTALLER_REQS = ["rich", "semver", "requests", "plumbum", "prompt-toolkit"]
@@ -21,40 +21,20 @@ OS = platform.uname().system
 ARCH = platform.uname().machine
 VERSION = "latest"
 
-### Feature flags
-# Install the virtualenv into the runtime dir
-FF_VENV_IN_RUNTIME = True
-
-# Install the wheel packaged with the installer
-FF_USE_LOCAL_WHEEL = True
-
 
 class Installer:
     """
     Deploys an InvokeAI installation into a given path
     """
 
+    reqs: list[str] = INSTALLER_REQS
+
     def __init__(self) -> None:
-        self.reqs = INSTALLER_REQS
-        self.preflight()
         if os.getenv("VIRTUAL_ENV") is not None:
             print("A virtual environment is already activated. Please 'deactivate' before installation.")
             sys.exit(-1)
         self.bootstrap()
-
-    def preflight(self) -> None:
-        """
-        Preflight checks
-        """
-
-        # TODO
-        # verify python version
-        # on macOS verify XCode tools are present
-        # verify libmesa, libglx on linux
-        # check that the system arch is not i386 (?)
-        # check that the system has a GPU, and the type of GPU
-
-        pass
+        self.available_releases = get_github_releases()
 
     def mktemp_venv(self) -> TemporaryDirectory:
         """
@@ -78,12 +58,9 @@ class Installer:
 
         return venv_dir
 
-    def bootstrap(self, verbose: bool = False) -> TemporaryDirectory:
+    def bootstrap(self, verbose: bool = False) -> TemporaryDirectory | None:
         """
         Bootstrap the installer venv with packages required at install time
-
-        :return: path to the virtual environment directory that was bootstrapped
-        :rtype: TemporaryDirectory
         """
 
         print("Initializing the installer. This may take a minute - please wait...")
@@ -95,39 +72,27 @@ class Installer:
         cmd.extend(self.reqs)
 
         try:
-            res = subprocess.check_output(cmd).decode()
+            # upgrade pip to the latest version to avoid a confusing message
+            res = upgrade_pip(Path(venv_dir.name))
             if verbose:
                 print(res)
+
+            # run the install prerequisites installation
+            res = subprocess.check_output(cmd).decode()
+
+            if verbose:
+                print(res)
+
             return venv_dir
         except subprocess.CalledProcessError as e:
             print(e)
 
-    def app_venv(self, path: str = None):
+    def app_venv(self, venv_parent) -> Path:
         """
         Create a virtualenv for the InvokeAI installation
         """
 
-        # explicit venv location
-        # currently unused in normal operation
-        # useful for testing or special cases
-        if path is not None:
-            venv_dir = Path(path)
-
-        # experimental / testing
-        elif not FF_VENV_IN_RUNTIME:
-            if OS == "Windows":
-                venv_dir_parent = os.getenv("APPDATA", "~/AppData/Roaming")
-            elif OS == "Darwin":
-                # there is no environment variable on macOS to find this
-                # TODO: confirm this is working as expected
-                venv_dir_parent = "~/Library/Application Support"
-            elif OS == "Linux":
-                venv_dir_parent = os.getenv("XDG_DATA_DIR", "~/.local/share")
-            venv_dir = Path(venv_dir_parent).expanduser().resolve() / f"InvokeAI/{VERSION}/venv"
-
-        # stable / current
-        else:
-            venv_dir = self.dest / ".venv"
+        venv_dir = venv_parent / ".venv"
 
         # Prefer to copy python executables
         # so that updates to system python don't break InvokeAI
@@ -141,7 +106,7 @@ class Installer:
         return venv_dir
 
     def install(
-        self, root: str = "~/invokeai", version: str = "latest", yes_to_all=False, find_links: Path = None
+        self, version=None, root: str = "~/invokeai", yes_to_all=False, find_links: Optional[Path] = None
     ) -> None:
         """
         Install the InvokeAI application into the given runtime path
@@ -158,15 +123,20 @@ class Installer:
 
         import messages
 
-        messages.welcome()
+        messages.welcome(self.available_releases)
 
-        default_path = os.environ.get("INVOKEAI_ROOT") or Path(root).expanduser().resolve()
-        self.dest = default_path if yes_to_all else messages.dest_path(root)
+        version = messages.choose_version(self.available_releases)
+
+        auto_dest = Path(os.environ.get("INVOKEAI_ROOT", root)).expanduser().resolve()
+        destination = auto_dest if yes_to_all else messages.dest_path(root)
+        if destination is None:
+            print("Could not find or create the destination directory. Installation cancelled.")
+            sys.exit(0)
 
         # create the venv for the app
-        self.venv = self.app_venv()
+        self.venv = self.app_venv(venv_parent=destination)
 
-        self.instance = InvokeAiInstance(runtime=self.dest, venv=self.venv, version=version)
+        self.instance = InvokeAiInstance(runtime=destination, venv=self.venv, version=version)
 
         # install dependencies and the InvokeAI application
         (extra_index_url, optional_modules) = get_torch_source() if not yes_to_all else (None, None)
@@ -190,7 +160,7 @@ class InvokeAiInstance:
     A single runtime directory *may* be shared by multiple virtual environments, though this isn't currently tested or supported.
     """
 
-    def __init__(self, runtime: Path, venv: Path, version: str) -> None:
+    def __init__(self, runtime: Path, venv: Path, version: str = "stable") -> None:
         self.runtime = runtime
         self.venv = venv
         self.pip = get_pip_from_venv(venv)
@@ -199,6 +169,7 @@ class InvokeAiInstance:
         set_sys_path(venv)
         os.environ["INVOKEAI_ROOT"] = str(self.runtime.expanduser().resolve())
         os.environ["VIRTUAL_ENV"] = str(self.venv.expanduser().resolve())
+        upgrade_pip(venv)
 
     def get(self) -> tuple[Path, Path]:
         """
@@ -212,54 +183,7 @@ class InvokeAiInstance:
 
     def install(self, extra_index_url=None, optional_modules=None, find_links=None):
         """
-        Install this instance, including dependencies and the app itself
-
-        :param extra_index_url: the "--extra-index-url ..." line for pip to look in extra indexes.
-        :type extra_index_url: str
-        """
-
-        import messages
-
-        # install torch first to ensure the correct version gets installed.
-        # works with either source or wheel install with negligible impact on installation times.
-        messages.simple_banner("Installing PyTorch :fire:")
-        self.install_torch(extra_index_url, find_links)
-
-        messages.simple_banner("Installing the InvokeAI Application :art:")
-        self.install_app(extra_index_url, optional_modules, find_links)
-
-    def install_torch(self, extra_index_url=None, find_links=None):
-        """
-        Install PyTorch
-        """
-
-        from plumbum import FG, local
-
-        pip = local[self.pip]
-
-        (
-            pip[
-                "install",
-                "--require-virtualenv",
-                "numpy~=1.24.0",  # choose versions that won't be uninstalled during phase 2
-                "urllib3~=1.26.0",
-                "requests~=2.28.0",
-                "torch==2.1.1",
-                "torchmetrics==0.11.4",
-                "torchvision>=0.16.1",
-                "--force-reinstall",
-                "--find-links" if find_links is not None else None,
-                find_links,
-                "--extra-index-url" if extra_index_url is not None else None,
-                extra_index_url,
-            ]
-            & FG
-        )
-
-    def install_app(self, extra_index_url=None, optional_modules=None, find_links=None):
-        """
-        Install the application with pip.
-        Supports installation from PyPi or from a local source directory.
+        Install the package from PyPi.
 
         :param extra_index_url: the "--extra-index-url ..." line for pip to look in extra indexes.
         :type extra_index_url: str
@@ -271,53 +195,52 @@ class InvokeAiInstance:
         :type find_links: Path
         """
 
-        ## this only applies to pypi installs; TODO actually use this
-        if self.version == "pre":
+        import messages
+
+        # not currently used, but may be useful for "install most recent version" option
+        if self.version == "prerelease":
             version = None
-            pre = "--pre"
+            pre_flag = "--pre"
+        elif self.version == "stable":
+            version = None
+            pre_flag = None
         else:
             version = self.version
-            pre = None
+            pre_flag = None
 
-        ## TODO: only local wheel will be installed as of now; support for --version arg is TODO
-        if FF_USE_LOCAL_WHEEL:
-            # if no wheel, try to do a source install before giving up
-            try:
-                src = str(next(Path(__file__).parent.glob("InvokeAI-*.whl")))
-            except StopIteration:
-                try:
-                    src = Path(__file__).parents[1].expanduser().resolve()
-                    # if the above directory contains one of these files, we'll do a source install
-                    next(src.glob("pyproject.toml"))
-                    next(src.glob("invokeai"))
-                except StopIteration:
-                    print("Unable to find a wheel or perform a source install. Giving up.")
+        src = "invokeai"
+        if optional_modules:
+            src += optional_modules
+        if version:
+            src += f"=={version}"
 
-        elif version == "source":
-            # this makes an assumption about the location of the installer package in the source tree
-            src = Path(__file__).parents[1].expanduser().resolve()
-        else:
-            # will install from PyPi
-            src = f"invokeai=={version}" if version is not None else "invokeai"
+        messages.simple_banner("Installing the InvokeAI Application :art:")
 
-        from plumbum import FG, local
+        from plumbum import FG, ProcessExecutionError, local  # type: ignore
 
         pip = local[self.pip]
 
-        (
-            pip[
-                "install",
-                "--require-virtualenv",
-                "--use-pep517",
-                str(src) + (optional_modules if optional_modules else ""),
-                "--find-links" if find_links is not None else None,
-                find_links,
-                "--extra-index-url" if extra_index_url is not None else None,
-                extra_index_url,
-                pre,
-            ]
-            & FG
-        )
+        pipeline = pip[
+            "install",
+            "--require-virtualenv",
+            "--force-reinstall",
+            "--use-pep517",
+            str(src),
+            "--find-links" if find_links is not None else None,
+            find_links,
+            "--extra-index-url" if extra_index_url is not None else None,
+            extra_index_url,
+            pre_flag,
+        ]
+
+        try:
+            _ = pipeline & FG
+        except ProcessExecutionError as e:
+            print(f"Error: {e}")
+            print(
+                "Could not install InvokeAI. Please try downloading the latest version of the installer and install again."
+            )
+            sys.exit(1)
 
     def configure(self):
         """
@@ -373,7 +296,6 @@ class InvokeAiInstance:
 
         ext = "bat" if OS == "Windows" else "sh"
 
-        # scripts = ['invoke', 'update']
         scripts = ["invoke"]
 
         for script in scripts:
@@ -408,6 +330,23 @@ def get_pip_from_venv(venv_path: Path) -> str:
     return str(venv_path.expanduser().resolve() / pip)
 
 
+def upgrade_pip(venv_path: Path) -> str | None:
+    """
+    Upgrade the pip executable in the given virtual environment
+    """
+
+    python = "Scripts\\python.exe" if OS == "Windows" else "bin/python"
+    python = str(venv_path.expanduser().resolve() / python)
+
+    try:
+        result = subprocess.check_output([python, "-m", "pip", "install", "--upgrade", "pip"]).decode()
+    except subprocess.CalledProcessError as e:
+        print(e)
+        result = None
+
+    return result
+
+
 def set_sys_path(venv_path: Path) -> None:
     """
     Given a path to a virtual environment, set the sys.path, in a cross-platform fashion,
@@ -431,7 +370,43 @@ def set_sys_path(venv_path: Path) -> None:
     sys.path.append(str(Path(venv_path, lib, "site-packages").expanduser().resolve()))
 
 
-def get_torch_source() -> (Union[str, None], str):
+def get_github_releases() -> tuple[list, list] | None:
+    """
+    Query Github for published (pre-)release versions.
+    Return a tuple where the first element is a list of stable releases and the second element is a list of pre-releases.
+    Return None if the query fails for any reason.
+    """
+
+    import requests
+
+    ## get latest releases using github api
+    url = "https://api.github.com/repos/invoke-ai/InvokeAI/releases"
+    releases, pre_releases = [], []
+    try:
+        res = requests.get(url)
+        res.raise_for_status()
+        tag_info = res.json()
+        for tag in tag_info:
+            if not tag["prerelease"]:
+                releases.append(tag["tag_name"].lstrip("v"))
+            else:
+                pre_releases.append(tag["tag_name"].lstrip("v"))
+    except requests.HTTPError as e:
+        print(f"Error: {e}")
+        print("Could not fetch version information from GitHub. Please check your network connection and try again.")
+        return
+    except Exception as e:
+        print(f"Error: {e}")
+        print("An unexpected error occurred while trying to fetch version information from GitHub. Please try again.")
+        return
+
+    releases.sort(reverse=True)
+    pre_releases.sort(reverse=True)
+
+    return releases, pre_releases
+
+
+def get_torch_source() -> Tuple[str | None, str | None]:
     """
     Determine the extra index URL for pip to use for torch installation.
     This depends on the OS and the graphics accelerator in use.
@@ -446,25 +421,26 @@ def get_torch_source() -> (Union[str, None], str):
     :rtype: list
     """
 
-    from messages import graphical_accelerator
+    from messages import select_gpu
 
-    # device can be one of: "cuda", "rocm", "cpu", "idk"
-    device = graphical_accelerator()
+    # device can be one of: "cuda", "rocm", "cpu", "cuda_and_dml, autodetect"
+    device = select_gpu()
 
     url = None
     optional_modules = "[onnx]"
     if OS == "Linux":
-        if device == "rocm":
-            url = "https://download.pytorch.org/whl/rocm5.4.2"
-        elif device == "cpu":
+        if device.value == "rocm":
+            url = "https://download.pytorch.org/whl/rocm5.6"
+        elif device.value == "cpu":
             url = "https://download.pytorch.org/whl/cpu"
 
-    if device == "cuda":
-        url = "https://download.pytorch.org/whl/cu121"
-        optional_modules = "[xformers,onnx-cuda]"
-    if device == "cuda_and_dml":
-        url = "https://download.pytorch.org/whl/cu121"
-        optional_modules = "[xformers,onnx-directml]"
+    elif OS == "Windows":
+        if device.value == "cuda":
+            url = "https://download.pytorch.org/whl/cu121"
+            optional_modules = "[xformers,onnx-cuda]"
+        if device.value == "cuda_and_dml":
+            url = "https://download.pytorch.org/whl/cu121"
+            optional_modules = "[xformers,onnx-directml]"
 
     # in all other cases, Torch wheels should be coming from PyPi as of Torch 1.13
 

installer/lib/messages.py

@@ -5,10 +5,11 @@ Installer user interaction
 
 import os
 import platform
+from enum import Enum
 from pathlib import Path
 
 from prompt_toolkit import HTML, prompt
-from prompt_toolkit.completion import PathCompleter
+from prompt_toolkit.completion import FuzzyWordCompleter, PathCompleter
 from prompt_toolkit.validation import Validator
 from rich import box, print
 from rich.console import Console, Group, group
@@ -35,16 +36,26 @@ else:
     console = Console(style=Style(color="grey74", bgcolor="grey19"))
 
 
-def welcome():
+def welcome(available_releases: tuple | None = None) -> None:
     @group()
     def text():
-        if (platform_specific := _platform_specific_help()) != "":
+        if (platform_specific := _platform_specific_help()) is not None:
             yield platform_specific
             yield ""
         yield Text.from_markup(
             "Some of the installation steps take a long time to run. Please be patient. If the script appears to hang for more than 10 minutes, please interrupt with [i]Control-C[/] and retry.",
             justify="center",
         )
+        if available_releases is not None:
+            latest_stable = available_releases[0][0]
+            last_pre = available_releases[1][0]
+            yield ""
+            yield Text.from_markup(
+                f"[red3]🠶[/] Latest stable release (recommended): [b bright_white]{latest_stable}", justify="center"
+            )
+            yield Text.from_markup(
+                f"[red3]🠶[/] Last published pre-release version: [b bright_white]{last_pre}", justify="center"
+            )
 
     console.rule()
     print(
@@ -61,19 +72,31 @@ def welcome():
     console.line()
 
 
-def confirm_install(dest: Path) -> bool:
-    if dest.exists():
-        print(f":exclamation: Directory {dest} already exists :exclamation:")
-        dest_confirmed = Confirm.ask(
-            ":stop_sign: (re)install in this location?",
-            default=False,
-        )
-    else:
-        print(f"InvokeAI will be installed in {dest}")
-        dest_confirmed = Confirm.ask("Use this location?", default=True)
+def choose_version(available_releases: tuple | None = None) -> str:
+    """
+    Prompt the user to choose an Invoke version to install
+    """
+
+    # short circuit if we couldn't get a version list
+    # still try to install the latest stable version
+    if available_releases is None:
+        return "stable"
+
+    console.print(":grey_question: [orange3]Please choose an Invoke version to install.")
+
+    choices = available_releases[0] + available_releases[1]
+
+    response = prompt(
+        message=f"   <Enter> to install the recommended release ({choices[0]}). <Tab> or type to pick a version: ",
+        complete_while_typing=True,
+        completer=FuzzyWordCompleter(choices),
+    )
+
+    console.print(f"   Version {choices[0] if response == "" else response} will be installed.")
+
     console.line()
 
-    return dest_confirmed
+    return "stable" if response == "" else response
 
 
 def user_wants_auto_configuration() -> bool:
@@ -109,7 +132,23 @@ def user_wants_auto_configuration() -> bool:
     return choice.lower().startswith("a")
 
 
-def dest_path(dest=None) -> Path:
+def confirm_install(dest: Path) -> bool:
+    if dest.exists():
+        print(f":stop_sign: Directory {dest} already exists!")
+        print("   Is this location correct?")
+        default = False
+    else:
+        print(f":file_folder: InvokeAI will be installed in {dest}")
+        default = True
+
+    dest_confirmed = Confirm.ask("   Please confirm:", default=default)
+
+    console.line()
+
+    return dest_confirmed
+
+
+def dest_path(dest=None) -> Path | None:
     """
     Prompt the user for the destination path and create the path
 
@@ -124,25 +163,21 @@ def dest_path(dest=None) -> Path:
     else:
         dest = Path.cwd().expanduser().resolve()
     prev_dest = init_path = dest
-
-    dest_confirmed = confirm_install(dest)
+    dest_confirmed = False
 
     while not dest_confirmed:
-        # if the given destination already exists, the starting point for browsing is its parent directory.
-        # the user may have made a typo, or otherwise wants to place the root dir next to an existing one.
-        # if the destination dir does NOT exist, then the user must have changed their mind about the selection.
-        # since we can't read their mind, start browsing at Path.cwd().
-        browse_start = (prev_dest.parent if prev_dest.exists() else Path.cwd()).expanduser().resolve()
+        browse_start = (dest or Path.cwd()).expanduser().resolve()
 
         path_completer = PathCompleter(
             only_directories=True,
             expanduser=True,
-            get_paths=lambda: [browse_start],  # noqa: B023
+            get_paths=lambda: [str(browse_start)],  # noqa: B023
             # get_paths=lambda: [".."].extend(list(browse_start.iterdir()))
         )
 
         console.line()
-        console.print(f"[orange3]Please select the destination directory for the installation:[/] \\[{browse_start}]: ")
+
+        console.print(f":grey_question: [orange3]Please select the install destination:[/] \\[{browse_start}]: ")
         selected = prompt(
             ">>> ",
             complete_in_thread=True,
@@ -155,6 +190,7 @@ def dest_path(dest=None) -> Path:
         )
         prev_dest = dest
         dest = Path(selected)
+
         console.line()
 
         dest_confirmed = confirm_install(dest.expanduser().resolve())
@@ -182,41 +218,45 @@ def dest_path(dest=None) -> Path:
         console.rule("Goodbye!")
 
 
-def graphical_accelerator():
+class GpuType(Enum):
+    CUDA = "cuda"
+    CUDA_AND_DML = "cuda_and_dml"
+    ROCM = "rocm"
+    CPU = "cpu"
+    AUTODETECT = "autodetect"
+
+
+def select_gpu() -> GpuType:
     """
-    Prompt the user to select the graphical accelerator in their system
-    This does not validate user's choices (yet), but only offers choices
-    valid for the platform.
-    CUDA is the fallback.
-    We may be able to detect the GPU driver by shelling out to `modprobe` or `lspci`,
-    but this is not yet supported or reliable. Also, some users may have exotic preferences.
+    Prompt the user to select the GPU driver
     """
 
     if ARCH == "arm64" and OS != "Darwin":
         print(f"Only CPU acceleration is available on {ARCH} architecture. Proceeding with that.")
-        return "cpu"
+        return GpuType.CPU
 
     nvidia = (
         "an [gold1 b]NVIDIA[/] GPU (using CUDA™)",
-        "cuda",
+        GpuType.CUDA,
     )
     nvidia_with_dml = (
         "an [gold1 b]NVIDIA[/] GPU (using CUDA™, and DirectML™ for ONNX) -- ALPHA",
-        "cuda_and_dml",
+        GpuType.CUDA_AND_DML,
     )
     amd = (
         "an [gold1 b]AMD[/] GPU (using ROCm™)",
-        "rocm",
+        GpuType.ROCM,
     )
     cpu = (
-        "no compatible GPU, or specifically prefer to use the CPU",
-        "cpu",
+        "Do not install any GPU support, use CPU for generation (slow)",
+        GpuType.CPU,
     )
-    idk = (
+    autodetect = (
         "I'm not sure what to choose",
-        "idk",
+        GpuType.AUTODETECT,
     )
 
+    options = []
     if OS == "Windows":
         options = [nvidia, nvidia_with_dml, cpu]
     if OS == "Linux":
@@ -230,7 +270,7 @@ def graphical_accelerator():
         return options[0][1]
 
     # "I don't know" is always added the last option
-    options.append(idk)
+    options.append(autodetect)  # type: ignore
 
     options = {str(i): opt for i, opt in enumerate(options, 1)}
 
@@ -265,9 +305,9 @@ def graphical_accelerator():
         ),
     )
 
-    if options[choice][1] == "idk":
+    if options[choice][1] is GpuType.AUTODETECT:
         console.print(
-            "No problem. We will try to install a version that [i]should[/i] be compatible. :crossed_fingers:"
+            "No problem. We will install CUDA support first :crossed_fingers: If Invoke does not detect a GPU, please re-run the installer and select one of the other GPU types."
         )
 
     return options[choice][1]
@@ -291,7 +331,7 @@ def windows_long_paths_registry() -> None:
     """
 
     with open(str(Path(__file__).parent / "WinLongPathsEnabled.reg"), "r", encoding="utf-16le") as code:
-        syntax = Syntax(code.read(), line_numbers=True)
+        syntax = Syntax(code.read(), line_numbers=True, lexer="regedit")
 
     console.print(
         Panel(
@@ -301,7 +341,7 @@ def windows_long_paths_registry() -> None:
                         "We will now apply a registry fix to enable long paths on Windows. InvokeAI needs this to function correctly. We are asking your permission to modify the Windows Registry on your behalf.",
                         "",
                         "This is the change that will be applied:",
-                        syntax,
+                        str(syntax),
                     ]
                 )
             ),
@@ -340,7 +380,7 @@ def introduction() -> None:
     console.line(2)
 
 
-def _platform_specific_help() -> str:
+def _platform_specific_help() -> Text | None:
     if OS == "Darwin":
         text = Text.from_markup(
             """[b wheat1]macOS Users![/]\n\nPlease be sure you have the [b wheat1]Xcode command-line tools[/] installed before continuing.\nIf not, cancel with [i]Control-C[/] and follow the Xcode install instructions at [deep_sky_blue1]https://www.freecodecamp.org/news/install-xcode-command-line-tools/[/]."""
@@ -354,5 +394,5 @@ def _platform_specific_help() -> str:
      [deep_sky_blue1]https://learn.microsoft.com/en-US/cpp/windows/latest-supported-vc-redist?view=msvc-170[/]"""
         )
     else:
-        text = ""
+        return
     return text

installer/templates/invoke.bat.in

@@ -15,7 +15,7 @@ echo 4. Download and install models
 echo 5. Change InvokeAI startup options
 echo 6. Re-run the configure script to fix a broken install or to complete a major upgrade
 echo 7. Open the developer console
-echo 8. Update InvokeAI
+echo 8. Update InvokeAI (DEPRECATED - please use the installer)
 echo 9. Run the InvokeAI image database maintenance script
 echo 10. Command-line help
 echo Q - Quit
@@ -52,8 +52,10 @@ IF /I "%choice%" == "1" (
     echo *** Type `exit` to quit this shell and deactivate the Python virtual environment ***
     call cmd /k
 ) ELSE IF /I "%choice%" == "8" (
-   echo Running invokeai-update...
-   python -m invokeai.frontend.install.invokeai_update
+    echo UPDATING FROM WITHIN THE APP IS BEING DEPRECATED.
+    echo Please download the installer from https://github.com/invoke-ai/InvokeAI/releases/latest and run it to update your installation.
+    timeout 4
+    python -m invokeai.frontend.install.invokeai_update
 ) ELSE IF /I "%choice%" == "9" (
    echo Running the db maintenance script...
    python .venv\Scripts\invokeai-db-maintenance.exe
@@ -77,4 +79,3 @@ pause
 
 :ending
 exit /b
-

installer/templates/invoke.sh.in

@@ -90,7 +90,9 @@ do_choice() {
         ;;
     8)
         clear
-        printf "Update InvokeAI\n"
+        printf "UPDATING FROM WITHIN THE APP IS BEING DEPRECATED\n"
+        printf "Please download the installer from https://github.com/invoke-ai/InvokeAI/releases/latest and run it to update your installation.\n"
+        sleep 4
         python -m invokeai.frontend.install.invokeai_update
         ;;
     9)
@@ -122,7 +124,7 @@ do_dialog() {
         5 "Change InvokeAI startup options"
         6 "Re-run the configure script to fix a broken install or to complete a major upgrade"
         7 "Open the developer console"
-        8 "Update InvokeAI"
+        8 "Update InvokeAI (DEPRECATED - please use the installer)"
 	9 "Run the InvokeAI image database maintenance script"
 	10 "Command-line help"
     )

installer/templates/update.bat.in

@@ -1,72 +0,0 @@
-@echo off
-setlocal EnableExtensions EnableDelayedExpansion
-
-PUSHD "%~dp0"
-
-set INVOKE_AI_VERSION=latest
-set arg=%1
-if "%arg%" neq "" (
-   if "%arg:~0,2%" equ "/?" (
-       echo Usage: update.bat ^<release name or branch^>
-       echo Updates InvokeAI to use the indicated version of the code base.
-       echo Find the version or branch for the release you want, and pass it as the argument.
-       echo For example '.\update.bat v2.2.5' for release 2.2.5.
-       echo             '.\update.bat main' for the latest development version
-       echo.
-       echo If no argument provided then will install the most recent release, equivalent to
-       echo '.\update.bat latest'
-       exit /b
-   ) else (
-       set INVOKE_AI_VERSION=%arg%
-   )
-)
-
-set INVOKE_AI_SRC="https://github.com/invoke-ai/InvokeAI/archive/!INVOKE_AI_VERSION!.zip"
-set INVOKE_AI_DEP=https://raw.githubusercontent.com/invoke-ai/InvokeAI/!INVOKE_AI_VERSION!/environments-and-requirements/requirements-base.txt
-set INVOKE_AI_MODELS=https://raw.githubusercontent.com/invoke-ai/InvokeAI/$INVOKE_AI_VERSION/configs/INITIAL_MODELS.yaml
-
-call curl -I "%INVOKE_AI_DEP%" -fs >.tmp.out
-if %errorlevel% neq 0 (
-    echo '!INVOKE_AI_VERSION!' is not a known branch name or tag. Please check the version and try again.
-    echo "Press any key to continue"
-    pause
-    exit /b
-)
-del .tmp.out
-
-echo This script will update InvokeAI and all its dependencies to !INVOKE_AI_SRC!.
-echo If you do not want to do this, press control-C now!
-pause
-
-call curl -L "%INVOKE_AI_DEP%" > environments-and-requirements/requirements-base.txt
-call curl -L "%INVOKE_AI_MODELS%" > configs/INITIAL_MODELS.yaml
-
-
-call .venv\Scripts\activate.bat
-call .venv\Scripts\python -mpip install -r requirements.txt
-if %errorlevel% neq 0 (
-   echo Installation of requirements failed. See https://invoke-ai.github.io/InvokeAI/installation/INSTALL_AUTOMATED/#troubleshooting for suggestions.
-   pause
-   exit /b
-)
-
-call .venv\Scripts\python -mpip install !INVOKE_AI_SRC!
-if %errorlevel% neq 0 (
-   echo Installation of InvokeAI failed. See https://invoke-ai.github.io/InvokeAI/installation/INSTALL_AUTOMATED/#troubleshooting for suggestions.
-   pause
-   exit /b
-)
-
-@rem call .venv\Scripts\invokeai-configure --root=.
-
-@rem if %errorlevel% neq 0 (
-@rem    echo Configuration InvokeAI failed. See https://invoke-ai.github.io/InvokeAI/installation/INSTALL_AUTOMATED/#troubleshooting for suggestions.
-@rem    pause
-@rem    exit /b
-@rem )
-
-echo InvokeAI has been updated to '%INVOKE_AI_VERSION%'
-
-echo "Press any key to continue"
-pause
-endlocal

installer/templates/update.sh.in

@@ -1,58 +0,0 @@
-#!/usr/bin/env bash
-
-set -eu
-
-if [ $# -ge 1 ] && [ "${1:0:2}" == "-h" ]; then
-    echo "Usage: update.sh <release>"
-    echo "Updates InvokeAI to use the indicated version of the code base."
-    echo "Find the version or branch for the release you want, and pass it as the argument."
-    echo "For example: update.sh v2.2.5 for release 2.2.5."
-    echo "             update.sh main for the current development version."
-    echo ""
-    echo "If no argument provided then will install the version tagged with 'latest', equivalent to"
-    echo "update.sh latest"
-    exit -1
-fi
-
-INVOKE_AI_VERSION=${1:-latest}
-
-INVOKE_AI_SRC="https://github.com/invoke-ai/InvokeAI/archive/$INVOKE_AI_VERSION.zip"
-INVOKE_AI_DEP=https://raw.githubusercontent.com/invoke-ai/InvokeAI/$INVOKE_AI_VERSION/environments-and-requirements/requirements-base.txt
-INVOKE_AI_MODELS=https://raw.githubusercontent.com/invoke-ai/InvokeAI/$INVOKE_AI_VERSION/configs/INITIAL_MODELS.yaml
-
-# ensure we're in the correct folder in case user's CWD is somewhere else
-scriptdir=$(dirname "$0")
-cd "$scriptdir"
-
-function _err_exit {
-    if test "$1" -ne 0
-    then
-	echo "Something went wrong while installing InvokeAI and/or its requirements."
-	echo "Update cannot continue. Please report this error to https://github.com/invoke-ai/InvokeAI/issues"
-        echo -e "Error code $1; Error caught was '$2'"
-        read -p "Press any key to exit..."
-        exit
-    fi
-}
-
-if ! curl -I "$INVOKE_AI_DEP" -fs >/dev/null; then
-    echo \'$INVOKE_AI_VERSION\' is not a known branch name or tag. Please check the version and try again.
-    exit
-fi
-
-echo This script will update InvokeAI and all its dependencies to version \'$INVOKE_AI_VERSION\'.
-echo If you do not want to do this, press control-C now!
-read -p "Press any key to continue, or CTRL-C to exit..."
-
-curl -L "$INVOKE_AI_DEP" > environments-and-requirements/requirements-base.txt
-curl -L "$INVOKE_AI_MODELS" > configs/INITIAL_MODELS.yaml
-
-. .venv/bin/activate
-
-./.venv/bin/python -mpip install -r requirements.txt
-_err_exit $? "The pip program failed to install InvokeAI's requirements."
-
-./.venv/bin/python -mpip install $INVOKE_AI_SRC
-_err_exit $? "The pip program failed to install InvokeAI."
-
-echo InvokeAI updated to \'$INVOKE_AI_VERSION\'

invokeai/app/api/dependencies.py

@@ -2,6 +2,7 @@
 
 from logging import Logger
 
+from invokeai.app.services.item_storage.item_storage_memory import ItemStorageMemory
 from invokeai.app.services.shared.sqlite.sqlite_util import init_db
 from invokeai.backend.util.logging import InvokeAILogger
 from invokeai.version.invokeai_version import __version__
@@ -11,6 +12,7 @@ from ..services.board_images.board_images_default import BoardImagesService
 from ..services.board_records.board_records_sqlite import SqliteBoardRecordStorage
 from ..services.boards.boards_default import BoardService
 from ..services.config import InvokeAIAppConfig
+from ..services.download import DownloadQueueService
 from ..services.image_files.image_files_disk import DiskImageFileStorage
 from ..services.image_records.image_records_sqlite import SqliteImageRecordStorage
 from ..services.images.images_default import ImageService
@@ -20,17 +22,15 @@ from ..services.invocation_queue.invocation_queue_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.item_storage.item_storage_sqlite import SqliteItemStorage
 from ..services.latents_storage.latents_storage_disk import DiskLatentsStorage
 from ..services.latents_storage.latents_storage_forward_cache import ForwardCacheLatentsStorage
-from ..services.model_install import ModelInstallService
 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
 from ..services.session_queue.session_queue_sqlite import SqliteSessionQueue
-from ..services.shared.default_graphs import create_system_graphs
-from ..services.shared.graph import GraphExecutionState, LibraryGraph
+from ..services.shared.graph import GraphExecutionState
 from ..services.urls.urls_default import LocalUrlService
 from ..services.workflow_records.workflow_records_sqlite import SqliteWorkflowRecordsStorage
 from .events import FastAPIEventService
@@ -61,7 +61,7 @@ class ApiDependencies:
     invoker: Invoker
 
     @staticmethod
-    def initialize(config: InvokeAIAppConfig, event_handler_id: int, logger: Logger = logger):
+    def initialize(config: InvokeAIAppConfig, event_handler_id: int, logger: Logger = logger) -> None:
         logger.info(f"InvokeAI version {__version__}")
         logger.info(f"Root directory = {str(config.root_path)}")
         logger.debug(f"Internet connectivity is {config.internet_available}")
@@ -79,16 +79,18 @@ class ApiDependencies:
         board_records = SqliteBoardRecordStorage(db=db)
         boards = BoardService()
         events = FastAPIEventService(event_handler_id)
-        graph_execution_manager = SqliteItemStorage[GraphExecutionState](db=db, table_name="graph_executions")
-        graph_library = SqliteItemStorage[LibraryGraph](db=db, table_name="graphs")
+        graph_execution_manager = ItemStorageMemory[GraphExecutionState]()
         image_records = SqliteImageRecordStorage(db=db)
         images = ImageService()
         invocation_cache = MemoryInvocationCache(max_cache_size=config.node_cache_size)
         latents = ForwardCacheLatentsStorage(DiskLatentsStorage(f"{output_folder}/latents"))
-        model_manager = ModelManagerService(config, logger)
-        model_record_service = ModelRecordServiceSQL(db=db)
-        model_install_service = ModelInstallService(
-            app_config=config, record_store=model_record_service, event_bus=events
+        download_queue_service = DownloadQueueService(event_bus=events)
+        model_metadata_service = ModelMetadataStoreSQL(db=db)
+        model_manager = ModelManagerService.build_model_manager(
+            app_config=configuration,
+            model_record_service=ModelRecordServiceSQL(db=db, metadata_store=model_metadata_service),
+            download_queue=download_queue_service,
+            events=events,
         )
         names = SimpleNameService()
         performance_statistics = InvocationStatsService()
@@ -107,7 +109,6 @@ class ApiDependencies:
             configuration=configuration,
             events=events,
             graph_execution_manager=graph_execution_manager,
-            graph_library=graph_library,
             image_files=image_files,
             image_records=image_records,
             images=images,
@@ -115,8 +116,7 @@ class ApiDependencies:
             latents=latents,
             logger=logger,
             model_manager=model_manager,
-            model_records=model_record_service,
-            model_install=model_install_service,
+            download_queue=download_queue_service,
             names=names,
             performance_statistics=performance_statistics,
             processor=processor,
@@ -127,12 +127,10 @@ class ApiDependencies:
             workflow_records=workflow_records,
         )
 
-        create_system_graphs(services.graph_library)
-
         ApiDependencies.invoker = Invoker(services)
         db.clean()
 
     @staticmethod
-    def shutdown():
+    def shutdown() -> None:
         if ApiDependencies.invoker:
             ApiDependencies.invoker.stop()

invokeai/app/api/no_cache_staticfiles.py

@@ -0,0 +1,28 @@
+from typing import Any
+
+from starlette.responses import Response
+from starlette.staticfiles import StaticFiles
+
+
+class NoCacheStaticFiles(StaticFiles):
+    """
+    This class is used to override the default caching behavior of starlette for static files,
+    ensuring we *never* cache static files. It modifies the file response headers to strictly
+    never cache the files.
+
+    Static files include the javascript bundles, fonts, locales, and some images. Generated
+    images are not included, as they are served by a router.
+    """
+
+    def __init__(self, *args: Any, **kwargs: Any):
+        self.cachecontrol = "max-age=0, no-cache, no-store, , must-revalidate"
+        self.pragma = "no-cache"
+        self.expires = "0"
+        super().__init__(*args, **kwargs)
+
+    def file_response(self, *args: Any, **kwargs: Any) -> Response:
+        resp = super().file_response(*args, **kwargs)
+        resp.headers.setdefault("Cache-Control", self.cachecontrol)
+        resp.headers.setdefault("Pragma", self.pragma)
+        resp.headers.setdefault("Expires", self.expires)
+        return resp

invokeai/app/api/routers/download_queue.py

@@ -0,0 +1,111 @@
+# Copyright (c) 2023 Lincoln D. Stein
+"""FastAPI route for the download queue."""
+
+from typing import List, Optional
+
+from fastapi import Body, Path, Response
+from fastapi.routing import APIRouter
+from pydantic.networks import AnyHttpUrl
+from starlette.exceptions import HTTPException
+
+from invokeai.app.services.download import (
+    DownloadJob,
+    UnknownJobIDException,
+)
+
+from ..dependencies import ApiDependencies
+
+download_queue_router = APIRouter(prefix="/v1/download_queue", tags=["download_queue"])
+
+
+@download_queue_router.get(
+    "/",
+    operation_id="list_downloads",
+)
+async def list_downloads() -> List[DownloadJob]:
+    """Get a list of active and inactive jobs."""
+    queue = ApiDependencies.invoker.services.download_queue
+    return queue.list_jobs()
+
+
+@download_queue_router.patch(
+    "/",
+    operation_id="prune_downloads",
+    responses={
+        204: {"description": "All completed jobs have been pruned"},
+        400: {"description": "Bad request"},
+    },
+)
+async def prune_downloads() -> Response:
+    """Prune completed and errored jobs."""
+    queue = ApiDependencies.invoker.services.download_queue
+    queue.prune_jobs()
+    return Response(status_code=204)
+
+
+@download_queue_router.post(
+    "/i/",
+    operation_id="download",
+)
+async def download(
+    source: AnyHttpUrl = Body(description="download source"),
+    dest: str = Body(description="download destination"),
+    priority: int = Body(default=10, description="queue priority"),
+    access_token: Optional[str] = Body(default=None, description="token for authorization to download"),
+) -> DownloadJob:
+    """Download the source URL to the file or directory indicted in dest."""
+    queue = ApiDependencies.invoker.services.download_queue
+    return queue.download(source, Path(dest), priority, access_token)
+
+
+@download_queue_router.get(
+    "/i/{id}",
+    operation_id="get_download_job",
+    responses={
+        200: {"description": "Success"},
+        404: {"description": "The requested download JobID could not be found"},
+    },
+)
+async def get_download_job(
+    id: int = Path(description="ID of the download job to fetch."),
+) -> DownloadJob:
+    """Get a download job using its ID."""
+    try:
+        job = ApiDependencies.invoker.services.download_queue.id_to_job(id)
+        return job
+    except UnknownJobIDException as e:
+        raise HTTPException(status_code=404, detail=str(e))
+
+
+@download_queue_router.delete(
+    "/i/{id}",
+    operation_id="cancel_download_job",
+    responses={
+        204: {"description": "Job has been cancelled"},
+        404: {"description": "The requested download JobID could not be found"},
+    },
+)
+async def cancel_download_job(
+    id: int = Path(description="ID of the download job to cancel."),
+) -> Response:
+    """Cancel a download job using its ID."""
+    try:
+        queue = ApiDependencies.invoker.services.download_queue
+        job = queue.id_to_job(id)
+        queue.cancel_job(job)
+        return Response(status_code=204)
+    except UnknownJobIDException as e:
+        raise HTTPException(status_code=404, detail=str(e))
+
+
+@download_queue_router.delete(
+    "/i",
+    operation_id="cancel_all_download_jobs",
+    responses={
+        204: {"description": "Download jobs have been cancelled"},
+    },
+)
+async def cancel_all_download_jobs() -> Response:
+    """Cancel all download jobs."""
+    ApiDependencies.invoker.services.download_queue.cancel_all_jobs()
+    return Response(status_code=204)

invokeai/app/api/routers/model_manager.py

@@ -0,0 +1,759 @@
+# Copyright (c) 2023 Lincoln D. Stein
+"""FastAPI route for model configuration records."""
+
+import pathlib
+import shutil
+from hashlib import sha1
+from random import randbytes
+from typing import Any, Dict, List, Optional, Set
+
+from fastapi import Body, Path, Query, Response
+from fastapi.routing import APIRouter
+from pydantic import BaseModel, ConfigDict
+from starlette.exceptions import HTTPException
+from typing_extensions import Annotated
+
+from invokeai.app.services.model_install import ModelInstallJob, ModelSource
+from invokeai.app.services.model_records import (
+    DuplicateModelException,
+    InvalidModelException,
+    ModelRecordOrderBy,
+    ModelSummary,
+    UnknownModelException,
+)
+from invokeai.app.services.shared.pagination import PaginatedResults
+from invokeai.backend.model_manager.config import (
+    AnyModelConfig,
+    BaseModelType,
+    MainCheckpointConfig,
+    ModelFormat,
+    ModelType,
+    SubModelType,
+)
+from invokeai.backend.model_manager.merge import MergeInterpolationMethod, ModelMerger
+from invokeai.backend.model_manager.metadata import AnyModelRepoMetadata
+
+from ..dependencies import ApiDependencies
+
+model_manager_router = APIRouter(prefix="/v2/models", tags=["model_manager"])
+
+
+class ModelsList(BaseModel):
+    """Return list of configs."""
+
+    models: List[AnyModelConfig]
+
+    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.
+##############################################################################
+example_model_config = {
+    "path": "string",
+    "name": "string",
+    "base": "sd-1",
+    "type": "main",
+    "format": "checkpoint",
+    "config": "string",
+    "key": "string",
+    "original_hash": "string",
+    "current_hash": "string",
+    "description": "string",
+    "source": "string",
+    "last_modified": 0,
+    "vae": "string",
+    "variant": "normal",
+    "prediction_type": "epsilon",
+    "repo_variant": "fp16",
+    "upcast_attention": False,
+    "ztsnr_training": False,
+}
+
+example_model_input = {
+    "path": "/path/to/model",
+    "name": "model_name",
+    "base": "sd-1",
+    "type": "main",
+    "format": "checkpoint",
+    "config": "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
+##############################################################################
+
+
+@model_manager_router.get(
+    "/",
+    operation_id="list_model_records",
+)
+async def list_model_records(
+    base_models: Optional[List[BaseModelType]] = Query(default=None, description="Base models to include"),
+    model_type: Optional[ModelType] = Query(default=None, description="The type of model to get"),
+    model_name: Optional[str] = Query(default=None, description="Exact match on the name of the model"),
+    model_format: Optional[ModelFormat] = Query(
+        default=None, description="Exact match on the format of the model (e.g. 'diffusers')"
+    ),
+) -> ModelsList:
+    """Get a list of models."""
+    record_store = ApiDependencies.invoker.services.model_manager.store
+    found_models: list[AnyModelConfig] = []
+    if base_models:
+        for base_model in base_models:
+            found_models.extend(
+                record_store.search_by_attr(
+                    base_model=base_model, model_type=model_type, model_name=model_name, model_format=model_format
+                )
+            )
+    else:
+        found_models.extend(
+            record_store.search_by_attr(model_type=model_type, model_name=model_name, model_format=model_format)
+        )
+    return ModelsList(models=found_models)
+
+
+@model_manager_router.get(
+    "/i/{key}",
+    operation_id="get_model_record",
+    responses={
+        200: {
+            "description": "The model configuration was retrieved successfully",
+            "content": {"application/json": {"example": example_model_config}},
+        },
+        400: {"description": "Bad request"},
+        404: {"description": "The model could not be found"},
+    },
+)
+async def get_model_record(
+    key: str = Path(description="Key of the model record to fetch."),
+) -> AnyModelConfig:
+    """Get a model record"""
+    record_store = ApiDependencies.invoker.services.model_manager.store
+    try:
+        config: AnyModelConfig = record_store.get_model(key)
+        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(
+    "/meta/i/{key}",
+    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"},
+        404: {"description": "No metadata available"},
+    },
+)
+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)
+    if not result:
+        raise HTTPException(status_code=404, detail="No metadata for a model with this 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(
+    "/tags/search",
+    operation_id="search_by_metadata_tags",
+)
+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)
+
+
+@model_manager_router.patch(
+    "/i/{key}",
+    operation_id="update_model_record",
+    responses={
+        200: {
+            "description": "The model was updated successfully",
+            "content": {"application/json": {"example": example_model_config}},
+        },
+        400: {"description": "Bad request"},
+        404: {"description": "The model could not be found"},
+        409: {"description": "There is already a model corresponding to the new name"},
+    },
+    status_code=200,
+)
+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)
+    ],
+) -> AnyModelConfig:
+    """Update model contents with a new config. If the model name or base fields are changed, then the model is renamed."""
+    logger = ApiDependencies.invoker.services.logger
+    record_store = ApiDependencies.invoker.services.model_manager.store
+    try:
+        model_response: AnyModelConfig = record_store.update_model(key, config=info)
+        logger.info(f"Updated model: {key}")
+    except UnknownModelException as e:
+        raise HTTPException(status_code=404, detail=str(e))
+    except ValueError as e:
+        logger.error(str(e))
+        raise HTTPException(status_code=409, detail=str(e))
+    return model_response
+
+
+@model_manager_router.delete(
+    "/i/{key}",
+    operation_id="del_model_record",
+    responses={
+        204: {"description": "Model deleted successfully"},
+        404: {"description": "Model not found"},
+    },
+    status_code=204,
+)
+async def del_model_record(
+    key: str = Path(description="Unique key of model to remove from model registry."),
+) -> Response:
+    """
+    Delete model record from database.
+
+    The configuration record will be removed. The corresponding weights files will be
+    deleted as well if they reside within the InvokeAI "models" directory.
+    """
+    logger = ApiDependencies.invoker.services.logger
+
+    try:
+        installer = ApiDependencies.invoker.services.model_manager.install
+        installer.delete(key)
+        logger.info(f"Deleted model: {key}")
+        return Response(status_code=204)
+    except UnknownModelException as e:
+        logger.error(str(e))
+        raise HTTPException(status_code=404, detail=str(e))
+
+
+@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
+    if config.key == "<NOKEY>":
+        config.key = sha1(randbytes(100)).hexdigest()
+        logger.info(f"Created model {config.key} for {config.name}")
+    try:
+        record_store.add_model(config.key, 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(
+    "/heuristic_import",
+    operation_id="heuristic_import_model",
+    responses={
+        201: {"description": "The model imported successfully"},
+        415: {"description": "Unrecognized file/folder format"},
+        424: {"description": "The model appeared to import successfully, but could not be found in the model manager"},
+        409: {"description": "There is already a model corresponding to this path or repo_id"},
+    },
+    status_code=201,
+)
+async def heuristic_import(
+    source: str,
+    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 ",
+        default=None,
+        example={"name": "modelT", "description": "antique cars"},
+    ),
+    access_token: Optional[str] = None,
+) -> ModelInstallJob:
+    """Install a model using a string identifier.
+
+    `source` can be any of the following.
+
+    1. A path on the local filesystem ('C:\\users\\fred\\model.safetensors')
+    2. A Url pointing to a single downloadable model file
+    3. A HuggingFace repo_id with any of the following formats:
+       - model/name
+       - model/name:fp16:vae
+       - model/name::vae          -- use default precision
+       - model/name:fp16:path/to/model.safetensors
+       - model/name::path/to/model.safetensors
+
+    `config` is an optional dict containing model configuration values that will override
+    the ones that are probed automatically.
+
+    `access_token` is an optional access token for use with Urls that require
+    authentication.
+
+    Models will be downloaded, probed, configured and installed in a
+    series of background threads. The return object has `status` attribute
+    that can be used to monitor progress.
+
+    See the documentation for `import_model_record` for more information on
+    interpreting the job information returned by this route.
+    """
+    logger = ApiDependencies.invoker.services.logger
+
+    try:
+        installer = ApiDependencies.invoker.services.model_manager.install
+        result: ModelInstallJob = installer.heuristic_import(
+            source=source,
+            config=config,
+        )
+        logger.info(f"Started installation of {source}")
+    except UnknownModelException as e:
+        logger.error(str(e))
+        raise HTTPException(status_code=424, detail=str(e))
+    except InvalidModelException as e:
+        logger.error(str(e))
+        raise HTTPException(status_code=415)
+    except ValueError as e:
+        logger.error(str(e))
+        raise HTTPException(status_code=409, detail=str(e))
+    return result
+
+
+@model_manager_router.post(
+    "/install",
+    operation_id="import_model",
+    responses={
+        201: {"description": "The model imported successfully"},
+        415: {"description": "Unrecognized file/folder format"},
+        424: {"description": "The model appeared to import successfully, but could not be found in the model manager"},
+        409: {"description": "There is already a model corresponding to this path or repo_id"},
+    },
+    status_code=201,
+)
+async def import_model(
+    source: ModelSource,
+    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 ",
+        default=None,
+    ),
+) -> ModelInstallJob:
+    """Install a model using its local path, repo_id, or remote URL.
+
+    Models will be downloaded, probed, configured and installed in a
+    series of background threads. The return object has `status` attribute
+    that can be used to monitor progress.
+
+    The source object is a discriminated Union of LocalModelSource,
+    HFModelSource and URLModelSource. Set the "type" field to the
+    appropriate value:
+
+    * To install a local path using LocalModelSource, pass a source of form:
+      ```
+      {
+        "type": "local",
+        "path": "/path/to/model",
+        "inplace": false
+      }
+      ```
+      The "inplace" flag, if true, will register the model in place in its
+      current filesystem location. Otherwise, the model will be copied
+      into the InvokeAI models directory.
+
+    * To install a HuggingFace repo_id using HFModelSource, pass a source of form:
+      ```
+      {
+        "type": "hf",
+        "repo_id": "stabilityai/stable-diffusion-2.0",
+        "variant": "fp16",
+        "subfolder": "vae",
+        "access_token": "f5820a918aaf01"
+      }
+      ```
+      The `variant`, `subfolder` and `access_token` fields are optional.
+
+    * To install a remote model using an arbitrary URL, pass:
+      ```
+      {
+        "type": "url",
+        "url": "http://www.civitai.com/models/123456",
+        "access_token": "f5820a918aaf01"
+      }
+      ```
+      The `access_token` field is optonal
+
+    The model's configuration record will be probed and filled in
+    automatically.  To override the default guesses, pass "metadata"
+    with a Dict containing the attributes you wish to override.
+
+    Installation occurs in the background. Either use list_model_install_jobs()
+    to poll for completion, or listen on the event bus for the following events:
+
+      * "model_install_running"
+      * "model_install_completed"
+      * "model_install_error"
+
+    On successful completion, the event's payload will contain the field "key"
+    containing the installed ID of the model. On an error, the event's payload
+    will contain the fields "error_type" and "error" describing the nature of the
+    error and its traceback, respectively.
+
+    """
+    logger = ApiDependencies.invoker.services.logger
+
+    try:
+        installer = ApiDependencies.invoker.services.model_manager.install
+        result: ModelInstallJob = installer.import_model(
+            source=source,
+            config=config,
+        )
+        logger.info(f"Started installation of {source}")
+    except UnknownModelException as e:
+        logger.error(str(e))
+        raise HTTPException(status_code=424, detail=str(e))
+    except InvalidModelException as e:
+        logger.error(str(e))
+        raise HTTPException(status_code=415)
+    except ValueError as e:
+        logger.error(str(e))
+        raise HTTPException(status_code=409, detail=str(e))
+    return result
+
+
+@model_manager_router.get(
+    "/import",
+    operation_id="list_model_install_jobs",
+)
+async def list_model_install_jobs() -> List[ModelInstallJob]:
+    """Return the list of model install jobs.
+
+    Install jobs have a numeric `id`, a `status`, and other fields that provide information on
+    the nature of the job and its progress. The `status` is one of:
+
+    * "waiting" -- Job is waiting in the queue to run
+    * "downloading" -- Model file(s) are downloading
+    * "running" -- Model has downloaded and the model probing and registration process is running
+    * "completed" -- Installation completed successfully
+    * "error" -- An error occurred. Details will be in the "error_type" and "error" fields.
+    * "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`.
+
+    See the example and schema below for more information.
+    """
+    jobs: List[ModelInstallJob] = ApiDependencies.invoker.services.model_manager.install.list_jobs()
+    return jobs
+
+
+@model_manager_router.get(
+    "/import/{id}",
+    operation_id="get_model_install_job",
+    responses={
+        200: {"description": "Success"},
+        404: {"description": "No such job"},
+    },
+)
+async def get_model_install_job(id: int = Path(description="Model install id")) -> ModelInstallJob:
+    """
+    Return model install job corresponding to the given source. See the documentation for 'List Model Install Jobs'
+    for information on the format of the return value.
+    """
+    try:
+        result: ModelInstallJob = ApiDependencies.invoker.services.model_manager.install.get_job_by_id(id)
+        return result
+    except ValueError as e:
+        raise HTTPException(status_code=404, detail=str(e))
+
+
+@model_manager_router.delete(
+    "/import/{id}",
+    operation_id="cancel_model_install_job",
+    responses={
+        201: {"description": "The job was cancelled successfully"},
+        415: {"description": "No such job"},
+    },
+    status_code=201,
+)
+async def cancel_model_install_job(id: int = Path(description="Model install job ID")) -> None:
+    """Cancel the model install job(s) corresponding to the given job ID."""
+    installer = ApiDependencies.invoker.services.model_manager.install
+    try:
+        job = installer.get_job_by_id(id)
+    except ValueError as e:
+        raise HTTPException(status_code=415, detail=str(e))
+    installer.cancel_job(job)
+
+
+@model_manager_router.patch(
+    "/import",
+    operation_id="prune_model_install_jobs",
+    responses={
+        204: {"description": "All completed and errored jobs have been pruned"},
+        400: {"description": "Bad request"},
+    },
+)
+async def prune_model_install_jobs() -> Response:
+    """Prune all completed and errored jobs from the install job list."""
+    ApiDependencies.invoker.services.model_manager.install.prune_jobs()
+    return Response(status_code=204)
+
+
+@model_manager_router.patch(
+    "/sync",
+    operation_id="sync_models_to_config",
+    responses={
+        204: {"description": "Model config record database resynced with files on disk"},
+        400: {"description": "Bad request"},
+    },
+)
+async def sync_models_to_config() -> Response:
+    """
+    Traverse the models and autoimport directories.
+
+    Model files without a corresponding
+    record in the database are added. Orphan records without a models file are deleted.
+    """
+    ApiDependencies.invoker.services.model_manager.install.sync_to_config()
+    return Response(status_code=204)
+
+
+@model_manager_router.put(
+    "/convert/{key}",
+    operation_id="convert_model",
+    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 convert_model(
+    key: str = Path(description="Unique key of the safetensors main model to convert to diffusers format."),
+) -> AnyModelConfig:
+    """
+    Permanently convert a model into diffusers format, replacing the safetensors version.
+    Note that during the conversion process the key and model hash will change.
+    The return value is the model configuration for the converted model.
+    """
+    logger = ApiDependencies.invoker.services.logger
+    loader = ApiDependencies.invoker.services.model_manager.load
+    store = ApiDependencies.invoker.services.model_manager.store
+    installer = ApiDependencies.invoker.services.model_manager.install
+
+    try:
+        model_config = store.get_model(key)
+    except UnknownModelException as e:
+        logger.error(str(e))
+        raise HTTPException(status_code=424, detail=str(e))
+
+    if not isinstance(model_config, MainCheckpointConfig):
+        logger.error(f"The model with key {key} is not a main checkpoint 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
+    loader.load_model_by_config(model_config, submodel_type=SubModelType.Scheduler)
+
+    # Get the path of the converted model from the loader
+    cache_path = loader.convert_cache.cache_path(key)
+    assert cache_path.exists()
+
+    # 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)
+
+    # install the diffusers
+    try:
+        new_key = installer.install_path(
+            cache_path,
+            config={
+                "name": original_name,
+                "description": model_config.description,
+                "original_hash": model_config.original_hash,
+                "source": model_config.source,
+            },
+        )
+    except DuplicateModelException as e:
+        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)
+
+    # delete the cached version
+    shutil.rmtree(cache_path)
+
+    # return the config record for the new diffusers directory
+    new_config: AnyModelConfig = store.get_model(new_key)
+    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

invokeai/app/api/routers/model_records.py

@@ -1,322 +0,0 @@
-# Copyright (c) 2023 Lincoln D. Stein
-"""FastAPI route for model configuration records."""
-
-
-from hashlib import sha1
-from random import randbytes
-from typing import Any, Dict, List, Optional
-
-from fastapi import Body, Path, Query, Response
-from fastapi.routing import APIRouter
-from pydantic import BaseModel, ConfigDict
-from starlette.exceptions import HTTPException
-from typing_extensions import Annotated
-
-from invokeai.app.services.model_install import ModelInstallJob, ModelSource
-from invokeai.app.services.model_records import (
-    DuplicateModelException,
-    InvalidModelException,
-    UnknownModelException,
-)
-from invokeai.backend.model_manager.config import (
-    AnyModelConfig,
-    BaseModelType,
-    ModelType,
-)
-
-from ..dependencies import ApiDependencies
-
-model_records_router = APIRouter(prefix="/v1/model/record", tags=["model_manager_v2"])
-
-
-class ModelsList(BaseModel):
-    """Return list of configs."""
-
-    models: list[AnyModelConfig]
-
-    model_config = ConfigDict(use_enum_values=True)
-
-
-@model_records_router.get(
-    "/",
-    operation_id="list_model_records",
-)
-async def list_model_records(
-    base_models: Optional[List[BaseModelType]] = Query(default=None, description="Base models to include"),
-    model_type: Optional[ModelType] = Query(default=None, description="The type of model to get"),
-    model_name: Optional[str] = Query(default=None, description="Exact match on the name of the model"),
-    model_format: Optional[str] = Query(
-        default=None, description="Exact match on the format of the model (e.g. 'diffusers')"
-    ),
-) -> ModelsList:
-    """Get a list of models."""
-    record_store = ApiDependencies.invoker.services.model_records
-    found_models: list[AnyModelConfig] = []
-    if base_models:
-        for base_model in base_models:
-            found_models.extend(
-                record_store.search_by_attr(
-                    base_model=base_model, model_type=model_type, model_name=model_name, model_format=model_format
-                )
-            )
-    else:
-        found_models.extend(
-            record_store.search_by_attr(model_type=model_type, model_name=model_name, model_format=model_format)
-        )
-    return ModelsList(models=found_models)
-
-
-@model_records_router.get(
-    "/i/{key}",
-    operation_id="get_model_record",
-    responses={
-        200: {"description": "Success"},
-        400: {"description": "Bad request"},
-        404: {"description": "The model could not be found"},
-    },
-)
-async def get_model_record(
-    key: str = Path(description="Key of the model record to fetch."),
-) -> AnyModelConfig:
-    """Get a model record"""
-    record_store = ApiDependencies.invoker.services.model_records
-    try:
-        return record_store.get_model(key)
-    except UnknownModelException as e:
-        raise HTTPException(status_code=404, detail=str(e))
-
-
-@model_records_router.patch(
-    "/i/{key}",
-    operation_id="update_model_record",
-    responses={
-        200: {"description": "The model was updated successfully"},
-        400: {"description": "Bad request"},
-        404: {"description": "The model could not be found"},
-        409: {"description": "There is already a model corresponding to the new name"},
-    },
-    status_code=200,
-    response_model=AnyModelConfig,
-)
-async def update_model_record(
-    key: Annotated[str, Path(description="Unique key of model")],
-    info: Annotated[AnyModelConfig, Body(description="Model config", discriminator="type")],
-) -> AnyModelConfig:
-    """Update model contents with a new config. If the model name or base fields are changed, then the model is renamed."""
-    logger = ApiDependencies.invoker.services.logger
-    record_store = ApiDependencies.invoker.services.model_records
-    try:
-        model_response = record_store.update_model(key, config=info)
-        logger.info(f"Updated model: {key}")
-    except UnknownModelException as e:
-        raise HTTPException(status_code=404, detail=str(e))
-    except ValueError as e:
-        logger.error(str(e))
-        raise HTTPException(status_code=409, detail=str(e))
-    return model_response
-
-
-@model_records_router.delete(
-    "/i/{key}",
-    operation_id="del_model_record",
-    responses={
-        204: {"description": "Model deleted successfully"},
-        404: {"description": "Model not found"},
-    },
-    status_code=204,
-)
-async def del_model_record(
-    key: str = Path(description="Unique key of model to remove from model registry."),
-) -> Response:
-    """
-    Delete model record from database.
-
-    The configuration record will be removed. The corresponding weights files will be
-    deleted as well if they reside within the InvokeAI "models" directory.
-    """
-    logger = ApiDependencies.invoker.services.logger
-
-    try:
-        installer = ApiDependencies.invoker.services.model_install
-        installer.delete(key)
-        logger.info(f"Deleted model: {key}")
-        return Response(status_code=204)
-    except UnknownModelException as e:
-        logger.error(str(e))
-        raise HTTPException(status_code=404, detail=str(e))
-
-
-@model_records_router.post(
-    "/i/",
-    operation_id="add_model_record",
-    responses={
-        201: {"description": "The model added successfully"},
-        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")],
-) -> AnyModelConfig:
-    """
-    Add a model using the configuration information appropriate for its type.
-    """
-    logger = ApiDependencies.invoker.services.logger
-    record_store = ApiDependencies.invoker.services.model_records
-    if config.key == "<NOKEY>":
-        config.key = sha1(randbytes(100)).hexdigest()
-        logger.info(f"Created model {config.key} for {config.name}")
-    try:
-        record_store.add_model(config.key, 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
-    return record_store.get_model(config.key)
-
-
-@model_records_router.post(
-    "/import",
-    operation_id="import_model_record",
-    responses={
-        201: {"description": "The model imported successfully"},
-        415: {"description": "Unrecognized file/folder format"},
-        424: {"description": "The model appeared to import successfully, but could not be found in the model manager"},
-        409: {"description": "There is already a model corresponding to this path or repo_id"},
-    },
-    status_code=201,
-)
-async def import_model(
-    source: ModelSource,
-    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 ",
-        default=None,
-    ),
-) -> ModelInstallJob:
-    """Add a model using its local path, repo_id, or remote URL.
-
-    Models will be downloaded, probed, configured and installed in a
-    series of background threads. The return object has `status` attribute
-    that can be used to monitor progress.
-
-    The source object is a discriminated Union of LocalModelSource,
-    HFModelSource and URLModelSource. Set the "type" field to the
-    appropriate value:
-
-    * To install a local path using LocalModelSource, pass a source of form:
-      `{
-        "type": "local",
-        "path": "/path/to/model",
-        "inplace": false
-      }`
-       The "inplace" flag, if true, will register the model in place in its
-       current filesystem location. Otherwise, the model will be copied
-       into the InvokeAI models directory.
-
-    * To install a HuggingFace repo_id using HFModelSource, pass a source of form:
-      `{
-        "type": "hf",
-        "repo_id": "stabilityai/stable-diffusion-2.0",
-        "variant": "fp16",
-        "subfolder": "vae",
-        "access_token": "f5820a918aaf01"
-      }`
-     The `variant`, `subfolder` and `access_token` fields are optional.
-
-    * To install a remote model using an arbitrary URL, pass:
-      `{
-        "type": "url",
-        "url": "http://www.civitai.com/models/123456",
-        "access_token": "f5820a918aaf01"
-      }`
-    The `access_token` field is optonal
-
-    The model's configuration record will be probed and filled in
-    automatically.  To override the default guesses, pass "metadata"
-    with a Dict containing the attributes you wish to override.
-
-    Installation occurs in the background. Either use list_model_install_jobs()
-    to poll for completion, or listen on the event bus for the following events:
-
-      "model_install_started"
-      "model_install_completed"
-      "model_install_error"
-
-    On successful completion, the event's payload will contain the field "key"
-    containing the installed ID of the model. On an error, the event's payload
-    will contain the fields "error_type" and "error" describing the nature of the
-    error and its traceback, respectively.
-
-    """
-    logger = ApiDependencies.invoker.services.logger
-
-    try:
-        installer = ApiDependencies.invoker.services.model_install
-        result: ModelInstallJob = installer.import_model(
-            source=source,
-            config=config,
-        )
-        logger.info(f"Started installation of {source}")
-    except UnknownModelException as e:
-        logger.error(str(e))
-        raise HTTPException(status_code=424, detail=str(e))
-    except InvalidModelException as e:
-        logger.error(str(e))
-        raise HTTPException(status_code=415)
-    except ValueError as e:
-        logger.error(str(e))
-        raise HTTPException(status_code=409, detail=str(e))
-    return result
-
-
-@model_records_router.get(
-    "/import",
-    operation_id="list_model_install_jobs",
-)
-async def list_model_install_jobs() -> List[ModelInstallJob]:
-    """
-    Return list of model install jobs.
-
-    If the optional 'source' argument is provided, then the list will be filtered
-    for partial string matches against the install source.
-    """
-    jobs: List[ModelInstallJob] = ApiDependencies.invoker.services.model_install.list_jobs()
-    return jobs
-
-
-@model_records_router.patch(
-    "/import",
-    operation_id="prune_model_install_jobs",
-    responses={
-        204: {"description": "All completed and errored jobs have been pruned"},
-        400: {"description": "Bad request"},
-    },
-)
-async def prune_model_install_jobs() -> Response:
-    """
-    Prune all completed and errored jobs from the install job list.
-    """
-    ApiDependencies.invoker.services.model_install.prune_jobs()
-    return Response(status_code=204)
-
-
-@model_records_router.patch(
-    "/sync",
-    operation_id="sync_models_to_config",
-    responses={
-        204: {"description": "Model config record database resynced with files on disk"},
-        400: {"description": "Bad request"},
-    },
-)
-async def sync_models_to_config() -> Response:
-    """
-    Traverse the models and autoimport directories. Model files without a corresponding
-    record in the database are added. Orphan records without a models file are deleted.
-    """
-    ApiDependencies.invoker.services.model_install.sync_to_config()
-    return Response(status_code=204)

invokeai/app/api/routers/models.py

@@ -1,427 +0,0 @@
-# Copyright (c) 2023 Kyle Schouviller (https://github.com/kyle0654), 2023 Kent Keirsey (https://github.com/hipsterusername), 2023 Lincoln D. Stein
-
-import pathlib
-from typing import Annotated, List, Literal, Optional, Union
-
-from fastapi import Body, Path, Query, Response
-from fastapi.routing import APIRouter
-from pydantic import BaseModel, ConfigDict, Field, TypeAdapter
-from starlette.exceptions import HTTPException
-
-from invokeai.backend import BaseModelType, ModelType
-from invokeai.backend.model_management import MergeInterpolationMethod
-from invokeai.backend.model_management.models import (
-    OPENAPI_MODEL_CONFIGS,
-    InvalidModelException,
-    ModelNotFoundException,
-    SchedulerPredictionType,
-)
-
-from ..dependencies import ApiDependencies
-
-models_router = APIRouter(prefix="/v1/models", tags=["models"])
-
-UpdateModelResponse = Union[tuple(OPENAPI_MODEL_CONFIGS)]
-UpdateModelResponseValidator = TypeAdapter(UpdateModelResponse)
-
-ImportModelResponse = Union[tuple(OPENAPI_MODEL_CONFIGS)]
-ImportModelResponseValidator = TypeAdapter(ImportModelResponse)
-
-ConvertModelResponse = Union[tuple(OPENAPI_MODEL_CONFIGS)]
-ConvertModelResponseValidator = TypeAdapter(ConvertModelResponse)
-
-MergeModelResponse = Union[tuple(OPENAPI_MODEL_CONFIGS)]
-ImportModelAttributes = Union[tuple(OPENAPI_MODEL_CONFIGS)]
-
-
-class ModelsList(BaseModel):
-    models: list[Union[tuple(OPENAPI_MODEL_CONFIGS)]]
-
-    model_config = ConfigDict(use_enum_values=True)
-
-
-ModelsListValidator = TypeAdapter(ModelsList)
-
-
-@models_router.get(
-    "/",
-    operation_id="list_models",
-    responses={200: {"model": ModelsList}},
-)
-async def list_models(
-    base_models: Optional[List[BaseModelType]] = Query(default=None, description="Base models to include"),
-    model_type: Optional[ModelType] = Query(default=None, description="The type of model to get"),
-) -> ModelsList:
-    """Gets a list of models"""
-    if base_models and len(base_models) > 0:
-        models_raw = []
-        for base_model in base_models:
-            models_raw.extend(ApiDependencies.invoker.services.model_manager.list_models(base_model, model_type))
-    else:
-        models_raw = ApiDependencies.invoker.services.model_manager.list_models(None, model_type)
-    models = ModelsListValidator.validate_python({"models": models_raw})
-    return models
-
-
-@models_router.patch(
-    "/{base_model}/{model_type}/{model_name}",
-    operation_id="update_model",
-    responses={
-        200: {"description": "The model was updated successfully"},
-        400: {"description": "Bad request"},
-        404: {"description": "The model could not be found"},
-        409: {"description": "There is already a model corresponding to the new name"},
-    },
-    status_code=200,
-    response_model=UpdateModelResponse,
-)
-async def update_model(
-    base_model: BaseModelType = Path(description="Base model"),
-    model_type: ModelType = Path(description="The type of model"),
-    model_name: str = Path(description="model name"),
-    info: Union[tuple(OPENAPI_MODEL_CONFIGS)] = Body(description="Model configuration"),
-) -> UpdateModelResponse:
-    """Update model contents with a new config. If the model name or base fields are changed, then the model is renamed."""
-    logger = ApiDependencies.invoker.services.logger
-
-    try:
-        previous_info = ApiDependencies.invoker.services.model_manager.list_model(
-            model_name=model_name,
-            base_model=base_model,
-            model_type=model_type,
-        )
-
-        # rename operation requested
-        if info.model_name != model_name or info.base_model != base_model:
-            ApiDependencies.invoker.services.model_manager.rename_model(
-                base_model=base_model,
-                model_type=model_type,
-                model_name=model_name,
-                new_name=info.model_name,
-                new_base=info.base_model,
-            )
-            logger.info(f"Successfully renamed {base_model.value}/{model_name}=>{info.base_model}/{info.model_name}")
-            # update information to support an update of attributes
-            model_name = info.model_name
-            base_model = info.base_model
-            new_info = ApiDependencies.invoker.services.model_manager.list_model(
-                model_name=model_name,
-                base_model=base_model,
-                model_type=model_type,
-            )
-            if new_info.get("path") != previous_info.get(
-                "path"
-            ):  # model manager moved model path during rename - don't overwrite it
-                info.path = new_info.get("path")
-
-        # replace empty string values with None/null to avoid phenomenon of vae: ''
-        info_dict = info.model_dump()
-        info_dict = {x: info_dict[x] if info_dict[x] else None for x in info_dict.keys()}
-
-        ApiDependencies.invoker.services.model_manager.update_model(
-            model_name=model_name,
-            base_model=base_model,
-            model_type=model_type,
-            model_attributes=info_dict,
-        )
-
-        model_raw = ApiDependencies.invoker.services.model_manager.list_model(
-            model_name=model_name,
-            base_model=base_model,
-            model_type=model_type,
-        )
-        model_response = UpdateModelResponseValidator.validate_python(model_raw)
-    except ModelNotFoundException as e:
-        raise HTTPException(status_code=404, detail=str(e))
-    except ValueError as e:
-        logger.error(str(e))
-        raise HTTPException(status_code=409, detail=str(e))
-    except Exception as e:
-        logger.error(str(e))
-        raise HTTPException(status_code=400, detail=str(e))
-
-    return model_response
-
-
-@models_router.post(
-    "/import",
-    operation_id="import_model",
-    responses={
-        201: {"description": "The model imported successfully"},
-        404: {"description": "The model could not be found"},
-        415: {"description": "Unrecognized file/folder format"},
-        424: {"description": "The model appeared to import successfully, but could not be found in the model manager"},
-        409: {"description": "There is already a model corresponding to this path or repo_id"},
-    },
-    status_code=201,
-    response_model=ImportModelResponse,
-)
-async def import_model(
-    location: str = Body(description="A model path, repo_id or URL to import"),
-    prediction_type: Optional[Literal["v_prediction", "epsilon", "sample"]] = Body(
-        description="Prediction type for SDv2 checkpoints and rare SDv1 checkpoints",
-        default=None,
-    ),
-) -> ImportModelResponse:
-    """Add a model using its local path, repo_id, or remote URL. Model characteristics will be probed and configured automatically"""
-
-    location = location.strip("\"' ")
-    items_to_import = {location}
-    prediction_types = {x.value: x for x in SchedulerPredictionType}
-    logger = ApiDependencies.invoker.services.logger
-
-    try:
-        installed_models = ApiDependencies.invoker.services.model_manager.heuristic_import(
-            items_to_import=items_to_import,
-            prediction_type_helper=lambda x: prediction_types.get(prediction_type),
-        )
-        info = installed_models.get(location)
-
-        if not info:
-            logger.error("Import failed")
-            raise HTTPException(status_code=415)
-
-        logger.info(f"Successfully imported {location}, got {info}")
-        model_raw = ApiDependencies.invoker.services.model_manager.list_model(
-            model_name=info.name, base_model=info.base_model, model_type=info.model_type
-        )
-        return ImportModelResponseValidator.validate_python(model_raw)
-
-    except ModelNotFoundException as e:
-        logger.error(str(e))
-        raise HTTPException(status_code=404, detail=str(e))
-    except InvalidModelException as e:
-        logger.error(str(e))
-        raise HTTPException(status_code=415)
-    except ValueError as e:
-        logger.error(str(e))
-        raise HTTPException(status_code=409, detail=str(e))
-
-
-@models_router.post(
-    "/add",
-    operation_id="add_model",
-    responses={
-        201: {"description": "The model added successfully"},
-        404: {"description": "The model could not be found"},
-        424: {"description": "The model appeared to add successfully, but could not be found in the model manager"},
-        409: {"description": "There is already a model corresponding to this path or repo_id"},
-    },
-    status_code=201,
-    response_model=ImportModelResponse,
-)
-async def add_model(
-    info: Union[tuple(OPENAPI_MODEL_CONFIGS)] = Body(description="Model configuration"),
-) -> ImportModelResponse:
-    """Add a model using the configuration information appropriate for its type. Only local models can be added by path"""
-
-    logger = ApiDependencies.invoker.services.logger
-
-    try:
-        ApiDependencies.invoker.services.model_manager.add_model(
-            info.model_name,
-            info.base_model,
-            info.model_type,
-            model_attributes=info.model_dump(),
-        )
-        logger.info(f"Successfully added {info.model_name}")
-        model_raw = ApiDependencies.invoker.services.model_manager.list_model(
-            model_name=info.model_name,
-            base_model=info.base_model,
-            model_type=info.model_type,
-        )
-        return ImportModelResponseValidator.validate_python(model_raw)
-    except ModelNotFoundException as e:
-        logger.error(str(e))
-        raise HTTPException(status_code=404, detail=str(e))
-    except ValueError as e:
-        logger.error(str(e))
-        raise HTTPException(status_code=409, detail=str(e))
-
-
-@models_router.delete(
-    "/{base_model}/{model_type}/{model_name}",
-    operation_id="del_model",
-    responses={
-        204: {"description": "Model deleted successfully"},
-        404: {"description": "Model not found"},
-    },
-    status_code=204,
-    response_model=None,
-)
-async def delete_model(
-    base_model: BaseModelType = Path(description="Base model"),
-    model_type: ModelType = Path(description="The type of model"),
-    model_name: str = Path(description="model name"),
-) -> Response:
-    """Delete Model"""
-    logger = ApiDependencies.invoker.services.logger
-
-    try:
-        ApiDependencies.invoker.services.model_manager.del_model(
-            model_name, base_model=base_model, model_type=model_type
-        )
-        logger.info(f"Deleted model: {model_name}")
-        return Response(status_code=204)
-    except ModelNotFoundException as e:
-        logger.error(str(e))
-        raise HTTPException(status_code=404, detail=str(e))
-
-
-@models_router.put(
-    "/convert/{base_model}/{model_type}/{model_name}",
-    operation_id="convert_model",
-    responses={
-        200: {"description": "Model converted successfully"},
-        400: {"description": "Bad request"},
-        404: {"description": "Model not found"},
-    },
-    status_code=200,
-    response_model=ConvertModelResponse,
-)
-async def convert_model(
-    base_model: BaseModelType = Path(description="Base model"),
-    model_type: ModelType = Path(description="The type of model"),
-    model_name: str = Path(description="model name"),
-    convert_dest_directory: Optional[str] = Query(
-        default=None, description="Save the converted model to the designated directory"
-    ),
-) -> ConvertModelResponse:
-    """Convert a checkpoint model into a diffusers model, optionally saving to the indicated destination directory, or `models` if none."""
-    logger = ApiDependencies.invoker.services.logger
-    try:
-        logger.info(f"Converting model: {model_name}")
-        dest = pathlib.Path(convert_dest_directory) if convert_dest_directory else None
-        ApiDependencies.invoker.services.model_manager.convert_model(
-            model_name,
-            base_model=base_model,
-            model_type=model_type,
-            convert_dest_directory=dest,
-        )
-        model_raw = ApiDependencies.invoker.services.model_manager.list_model(
-            model_name, base_model=base_model, model_type=model_type
-        )
-        response = ConvertModelResponseValidator.validate_python(model_raw)
-    except ModelNotFoundException as e:
-        raise HTTPException(status_code=404, detail=f"Model '{model_name}' not found: {str(e)}")
-    except ValueError as e:
-        raise HTTPException(status_code=400, detail=str(e))
-    return response
-
-
-@models_router.get(
-    "/search",
-    operation_id="search_for_models",
-    responses={
-        200: {"description": "Directory searched successfully"},
-        404: {"description": "Invalid directory path"},
-    },
-    status_code=200,
-    response_model=List[pathlib.Path],
-)
-async def search_for_models(
-    search_path: pathlib.Path = Query(description="Directory path to search for models"),
-) -> List[pathlib.Path]:
-    if not search_path.is_dir():
-        raise HTTPException(
-            status_code=404,
-            detail=f"The search path '{search_path}' does not exist or is not directory",
-        )
-    return ApiDependencies.invoker.services.model_manager.search_for_models(search_path)
-
-
-@models_router.get(
-    "/ckpt_confs",
-    operation_id="list_ckpt_configs",
-    responses={
-        200: {"description": "paths retrieved successfully"},
-    },
-    status_code=200,
-    response_model=List[pathlib.Path],
-)
-async def list_ckpt_configs() -> List[pathlib.Path]:
-    """Return a list of the legacy checkpoint configuration files stored in `ROOT/configs/stable-diffusion`, relative to ROOT."""
-    return ApiDependencies.invoker.services.model_manager.list_checkpoint_configs()
-
-
-@models_router.post(
-    "/sync",
-    operation_id="sync_to_config",
-    responses={
-        201: {"description": "synchronization successful"},
-    },
-    status_code=201,
-    response_model=bool,
-)
-async def sync_to_config() -> bool:
-    """Call after making changes to models.yaml, autoimport directories or models directory to synchronize
-    in-memory data structures with disk data structures."""
-    ApiDependencies.invoker.services.model_manager.sync_to_config()
-    return True
-
-
-# There's some weird pydantic-fastapi behaviour that requires this to be a separate class
-# TODO: After a few updates, see if it works inside the route operation handler?
-class MergeModelsBody(BaseModel):
-    model_names: List[str] = Field(description="model name", min_length=2, max_length=3)
-    merged_model_name: Optional[str] = Field(description="Name of destination model")
-    alpha: Optional[float] = Field(description="Alpha weighting strength to apply to 2d and 3d models", default=0.5)
-    interp: Optional[MergeInterpolationMethod] = Field(description="Interpolation method")
-    force: Optional[bool] = Field(
-        description="Force merging of models created with different versions of diffusers",
-        default=False,
-    )
-
-    merge_dest_directory: Optional[str] = Field(
-        description="Save the merged model to the designated directory (with 'merged_model_name' appended)",
-        default=None,
-    )
-
-    model_config = ConfigDict(protected_namespaces=())
-
-
-@models_router.put(
-    "/merge/{base_model}",
-    operation_id="merge_models",
-    responses={
-        200: {"description": "Model converted successfully"},
-        400: {"description": "Incompatible models"},
-        404: {"description": "One or more models not found"},
-    },
-    status_code=200,
-    response_model=MergeModelResponse,
-)
-async def merge_models(
-    body: Annotated[MergeModelsBody, Body(description="Model configuration", embed=True)],
-    base_model: BaseModelType = Path(description="Base model"),
-) -> MergeModelResponse:
-    """Convert a checkpoint model into a diffusers model"""
-    logger = ApiDependencies.invoker.services.logger
-    try:
-        logger.info(
-            f"Merging models: {body.model_names} into {body.merge_dest_directory or '<MODELS>'}/{body.merged_model_name}"
-        )
-        dest = pathlib.Path(body.merge_dest_directory) if body.merge_dest_directory else None
-        result = ApiDependencies.invoker.services.model_manager.merge_models(
-            model_names=body.model_names,
-            base_model=base_model,
-            merged_model_name=body.merged_model_name or "+".join(body.model_names),
-            alpha=body.alpha,
-            interp=body.interp,
-            force=body.force,
-            merge_dest_directory=dest,
-        )
-        model_raw = ApiDependencies.invoker.services.model_manager.list_model(
-            result.name,
-            base_model=base_model,
-            model_type=ModelType.Main,
-        )
-        response = ConvertModelResponseValidator.validate_python(model_raw)
-    except ModelNotFoundException:
-        raise HTTPException(
-            status_code=404,
-            detail=f"One or more of the models '{body.model_names}' not found",
-        )
-    except ValueError as e:
-        raise HTTPException(status_code=400, detail=str(e))
-    return response

invokeai/app/api/routers/utilities.py

@@ -23,10 +23,11 @@ class DynamicPromptsResponse(BaseModel):
 )
 async def parse_dynamicprompts(
     prompt: str = Body(description="The prompt to parse with dynamicprompts"),
-    max_prompts: int = Body(default=1000, description="The max number of prompts to generate"),
+    max_prompts: int = Body(ge=1, le=10000, default=1000, description="The max number of prompts to generate"),
     combinatorial: bool = Body(default=True, description="Whether to use the combinatorial generator"),
 ) -> DynamicPromptsResponse:
     """Creates a batch process"""
+    max_prompts = min(max_prompts, 10000)
     generator: Union[RandomPromptGenerator, CombinatorialPromptGenerator]
     try:
         error: Optional[str] = None

invokeai/app/api/sockets.py

@@ -14,7 +14,7 @@ class SocketIO:
 
     def __init__(self, app: FastAPI):
         self.__sio = AsyncServer(async_mode="asgi", cors_allowed_origins="*")
-        self.__app = ASGIApp(socketio_server=self.__sio, socketio_path="socket.io")
+        self.__app = ASGIApp(socketio_server=self.__sio, socketio_path="/ws/socket.io")
         app.mount("/ws", self.__app)
 
         self.__sio.on("subscribe_queue", handler=self._handle_sub_queue)

invokeai/app/api_app.py

@@ -3,6 +3,7 @@
 # values from the command line or config file.
 import sys
 
+from invokeai.app.api.no_cache_staticfiles import NoCacheStaticFiles
 from invokeai.version.invokeai_version import __version__
 
 from .services.config import InvokeAIAppConfig
@@ -27,8 +28,7 @@ if True:  # hack to make flake8 happy with imports coming after setting up the c
     from fastapi.middleware.gzip import GZipMiddleware
     from fastapi.openapi.docs import get_redoc_html, get_swagger_ui_html
     from fastapi.openapi.utils import get_openapi
-    from fastapi.responses import FileResponse, HTMLResponse
-    from fastapi.staticfiles import StaticFiles
+    from fastapi.responses import HTMLResponse
     from fastapi_events.handlers.local import local_handler
     from fastapi_events.middleware import EventHandlerASGIMiddleware
     from pydantic.json_schema import models_json_schema
@@ -45,9 +45,9 @@ if True:  # hack to make flake8 happy with imports coming after setting up the c
         app_info,
         board_images,
         boards,
+        download_queue,
         images,
-        model_records,
-        models,
+        model_manager,
         session_queue,
         sessions,
         utilities,
@@ -75,7 +75,7 @@ mimetypes.add_type("text/css", ".css")
 
 # Create the app
 # TODO: create this all in a method so configuration/etc. can be passed in?
-app = FastAPI(title="Invoke AI", 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)
 
 # Add event handler
 event_handler_id: int = id(app)
@@ -114,8 +114,8 @@ async def shutdown_event() -> None:
 app.include_router(sessions.session_router, prefix="/api")
 
 app.include_router(utilities.utilities_router, prefix="/api")
-app.include_router(models.models_router, prefix="/api")
-app.include_router(model_records.model_records_router, prefix="/api")
+app.include_router(model_manager.model_manager_router, prefix="/api")
+app.include_router(download_queue.download_queue_router, prefix="/api")
 app.include_router(images.images_router, prefix="/api")
 app.include_router(boards.boards_router, prefix="/api")
 app.include_router(board_images.board_images_router, prefix="/api")
@@ -176,21 +176,23 @@ def custom_openapi() -> dict[str, Any]:
         invoker_schema["class"] = "invocation"
         openapi_schema["components"]["schemas"][f"{output_type_title}"]["class"] = "output"
 
-    from invokeai.backend.model_management.models import get_model_config_enums
+    # This code no longer seems to be necessary?
+    # Leave it here just in case
+    #
+    # from invokeai.backend.model_manager import get_model_config_formats
+    # formats = get_model_config_formats()
+    # for model_config_name, enum_set in formats.items():
 
-    for model_config_format_enum in set(get_model_config_enums()):
-        name = model_config_format_enum.__qualname__
+    #     if model_config_name in openapi_schema["components"]["schemas"]:
+    #         # print(f"Config with name {name} already defined")
+    #         continue
 
-        if name in openapi_schema["components"]["schemas"]:
-            # print(f"Config with name {name} already defined")
-            continue
-
-        openapi_schema["components"]["schemas"][name] = {
-            "title": name,
-            "description": "An enumeration.",
-            "type": "string",
-            "enum": [v.value for v in model_config_format_enum],
-        }
+    #     openapi_schema["components"]["schemas"][model_config_name] = {
+    #         "title": model_config_name,
+    #         "description": "An enumeration.",
+    #         "type": "string",
+    #         "enum": [v.value for v in enum_set],
+    #     }
 
     app.openapi_schema = openapi_schema
     return app.openapi_schema
@@ -203,8 +205,8 @@ app.openapi = custom_openapi  # type: ignore [method-assign] # this is a valid a
 def overridden_swagger() -> HTMLResponse:
     return get_swagger_ui_html(
         openapi_url=app.openapi_url,  # type: ignore [arg-type] # this is always a string
-        title=app.title,
-        swagger_favicon_url="/static/docs/favicon.ico",
+        title=f"{app.title} - Swagger UI",
+        swagger_favicon_url="static/docs/invoke-favicon-docs.svg",
     )
 
 
@@ -212,26 +214,20 @@ def overridden_swagger() -> HTMLResponse:
 def overridden_redoc() -> HTMLResponse:
     return get_redoc_html(
         openapi_url=app.openapi_url,  # type: ignore [arg-type] # this is always a string
-        title=app.title,
-        redoc_favicon_url="/static/docs/favicon.ico",
+        title=f"{app.title} - Redoc",
+        redoc_favicon_url="static/docs/invoke-favicon-docs.svg",
     )
 
 
 web_root_path = Path(list(web_dir.__path__)[0])
 
-# Only serve the UI if we it has a build
-if (web_root_path / "dist").exists():
-    # Cannot add headers to StaticFiles, so we must serve index.html with a custom route
-    # Add cache-control: no-store header to prevent caching of index.html, which leads to broken UIs at release
-    @app.get("/", include_in_schema=False, name="ui_root")
-    def get_index() -> FileResponse:
-        return FileResponse(Path(web_root_path, "dist/index.html"), headers={"Cache-Control": "no-store"})
-
-    # # Must mount *after* the other routes else it borks em
-    app.mount("/assets", StaticFiles(directory=Path(web_root_path, "dist/assets/")), name="assets")
-    app.mount("/locales", StaticFiles(directory=Path(web_root_path, "dist/locales/")), name="locales")
-
-app.mount("/static", StaticFiles(directory=Path(web_root_path, "static/")), name="static")  # docs favicon is in here
+try:
+    app.mount("/", NoCacheStaticFiles(directory=Path(web_root_path, "dist"), html=True), name="ui")
+except RuntimeError:
+    logger.warn(f"No UI found at {web_root_path}/dist, skipping UI mount")
+app.mount(
+    "/static", NoCacheStaticFiles(directory=Path(web_root_path, "static/")), name="static"
+)  # docs favicon is in here
 
 
 def invoke_api() -> None:

invokeai/app/invocations/compel.py

@@ -1,22 +1,27 @@
-import re
 from dataclasses import dataclass
-from typing import List, Optional, Union
+from typing import Iterator, List, Optional, Tuple, Union
 
 import torch
 from compel import Compel, ReturnedEmbeddingsType
 from compel.prompt_parser import Blend, Conjunction, CrossAttentionControlSubstitute, FlattenedPrompt, Fragment
+from transformers import CLIPTokenizer
 
+import invokeai.backend.util.logging as logger
 from invokeai.app.invocations.primitives import ConditioningField, ConditioningOutput
+from invokeai.app.services.model_records import UnknownModelException
 from invokeai.app.shared.fields import FieldDescriptions
+from invokeai.app.util.ti_utils import extract_ti_triggers_from_prompt
+from invokeai.backend.lora import LoRAModelRaw
+from invokeai.backend.model_manager import ModelType
+from invokeai.backend.model_patcher import ModelPatcher
 from invokeai.backend.stable_diffusion.diffusion.conditioning_data import (
     BasicConditioningInfo,
     ExtraConditioningInfo,
     SDXLConditioningInfo,
 )
+from invokeai.backend.textual_inversion import TextualInversionModelRaw
+from invokeai.backend.util.devices import torch_dtype
 
-from ...backend.model_management.lora import ModelPatcher
-from ...backend.model_management.models import ModelNotFoundException, ModelType
-from ...backend.util.devices import torch_dtype
 from .baseinvocation import (
     BaseInvocation,
     BaseInvocationOutput,
@@ -66,49 +71,45 @@ class CompelInvocation(BaseInvocation):
 
     @torch.no_grad()
     def invoke(self, context: InvocationContext) -> ConditioningOutput:
-        tokenizer_info = context.services.model_manager.get_model(
+        tokenizer_info = context.services.model_manager.load_model_by_key(
             **self.clip.tokenizer.model_dump(),
             context=context,
         )
-        text_encoder_info = context.services.model_manager.get_model(
+        text_encoder_info = context.services.model_manager.load_model_by_key(
             **self.clip.text_encoder.model_dump(),
             context=context,
         )
 
-        def _lora_loader():
+        def _lora_loader() -> Iterator[Tuple[LoRAModelRaw, float]]:
             for lora in self.clip.loras:
-                lora_info = context.services.model_manager.get_model(
+                lora_info = context.services.model_manager.load_model_by_key(
                     **lora.model_dump(exclude={"weight"}), context=context
                 )
-                yield (lora_info.context.model, lora.weight)
+                assert isinstance(lora_info.model, LoRAModelRaw)
+                yield (lora_info.model, lora.weight)
                 del lora_info
             return
 
         # loras = [(context.services.model_manager.get_model(**lora.dict(exclude={"weight"})).context.model, lora.weight) for lora in self.clip.loras]
 
         ti_list = []
-        for trigger in re.findall(r"<[a-zA-Z0-9., _-]+>", self.prompt):
+        for trigger in extract_ti_triggers_from_prompt(self.prompt):
             name = trigger[1:-1]
             try:
-                ti_list.append(
-                    (
-                        name,
-                        context.services.model_manager.get_model(
-                            model_name=name,
-                            base_model=self.clip.text_encoder.base_model,
-                            model_type=ModelType.TextualInversion,
-                            context=context,
-                        ).context.model,
-                    )
-                )
-            except ModelNotFoundException:
+                loaded_model = context.services.model_manager.load_model_by_key(
+                    **self.clip.text_encoder.model_dump(),
+                    context=context,
+                ).model
+                assert isinstance(loaded_model, TextualInversionModelRaw)
+                ti_list.append((name, loaded_model))
+            except UnknownModelException:
                 # print(e)
                 # import traceback
                 # print(traceback.format_exc())
                 print(f'Warn: trigger: "{trigger}" not found')
 
         with (
-            ModelPatcher.apply_ti(tokenizer_info.context.model, text_encoder_info.context.model, ti_list) as (
+            ModelPatcher.apply_ti(tokenizer_info.model, text_encoder_info.model, ti_list) as (
                 tokenizer,
                 ti_manager,
             ),
@@ -116,7 +117,7 @@ class CompelInvocation(BaseInvocation):
             # Apply the LoRA after text_encoder has been moved to its target device for faster patching.
             ModelPatcher.apply_lora_text_encoder(text_encoder, _lora_loader()),
             # Apply CLIP Skip after LoRA to prevent LoRA application from failing on skipped layers.
-            ModelPatcher.apply_clip_skip(text_encoder_info.context.model, self.clip.skipped_layers),
+            ModelPatcher.apply_clip_skip(text_encoder_info.model, self.clip.skipped_layers),
         ):
             compel = Compel(
                 tokenizer=tokenizer,
@@ -150,7 +151,7 @@ class CompelInvocation(BaseInvocation):
         )
 
         conditioning_name = f"{context.graph_execution_state_id}_{self.id}_conditioning"
-        context.services.latents.save(conditioning_name, conditioning_data)
+        context.services.latents.save(conditioning_name, conditioning_data)  # TODO: fix type mismatch here
 
         return ConditioningOutput(
             conditioning=ConditioningField(
@@ -160,6 +161,8 @@ class CompelInvocation(BaseInvocation):
 
 
 class SDXLPromptInvocationBase:
+    """Prompt processor for SDXL models."""
+
     def run_clip_compel(
         self,
         context: InvocationContext,
@@ -168,26 +171,27 @@ class SDXLPromptInvocationBase:
         get_pooled: bool,
         lora_prefix: str,
         zero_on_empty: bool,
-    ):
-        tokenizer_info = context.services.model_manager.get_model(
+    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[ExtraConditioningInfo]]:
+        tokenizer_info = context.services.model_manager.load_model_by_key(
             **clip_field.tokenizer.model_dump(),
             context=context,
         )
-        text_encoder_info = context.services.model_manager.get_model(
+        text_encoder_info = context.services.model_manager.load_model_by_key(
             **clip_field.text_encoder.model_dump(),
             context=context,
         )
 
         # return zero on empty
         if prompt == "" and zero_on_empty:
-            cpu_text_encoder = text_encoder_info.context.model
+            cpu_text_encoder = text_encoder_info.model
+            assert isinstance(cpu_text_encoder, torch.nn.Module)
             c = torch.zeros(
                 (
                     1,
                     cpu_text_encoder.config.max_position_embeddings,
                     cpu_text_encoder.config.hidden_size,
                 ),
-                dtype=text_encoder_info.context.cache.precision,
+                dtype=cpu_text_encoder.dtype,
             )
             if get_pooled:
                 c_pooled = torch.zeros(
@@ -198,40 +202,41 @@ class SDXLPromptInvocationBase:
                 c_pooled = None
             return c, c_pooled, None
 
-        def _lora_loader():
+        def _lora_loader() -> Iterator[Tuple[LoRAModelRaw, float]]:
             for lora in clip_field.loras:
-                lora_info = context.services.model_manager.get_model(
+                lora_info = context.services.model_manager.load_model_by_key(
                     **lora.model_dump(exclude={"weight"}), context=context
                 )
-                yield (lora_info.context.model, lora.weight)
+                lora_model = lora_info.model
+                assert isinstance(lora_model, LoRAModelRaw)
+                yield (lora_model, lora.weight)
                 del lora_info
             return
 
         # loras = [(context.services.model_manager.get_model(**lora.dict(exclude={"weight"})).context.model, lora.weight) for lora in self.clip.loras]
 
         ti_list = []
-        for trigger in re.findall(r"<[a-zA-Z0-9., _-]+>", prompt):
+        for trigger in extract_ti_triggers_from_prompt(prompt):
             name = trigger[1:-1]
             try:
-                ti_list.append(
-                    (
-                        name,
-                        context.services.model_manager.get_model(
-                            model_name=name,
-                            base_model=clip_field.text_encoder.base_model,
-                            model_type=ModelType.TextualInversion,
-                            context=context,
-                        ).context.model,
-                    )
-                )
-            except ModelNotFoundException:
+                ti_model = context.services.model_manager.load_model_by_attr(
+                    model_name=name,
+                    base_model=text_encoder_info.config.base,
+                    model_type=ModelType.TextualInversion,
+                    context=context,
+                ).model
+                assert isinstance(ti_model, TextualInversionModelRaw)
+                ti_list.append((name, ti_model))
+            except UnknownModelException:
                 # print(e)
                 # import traceback
                 # print(traceback.format_exc())
-                print(f'Warn: trigger: "{trigger}" not found')
+                logger.warning(f'trigger: "{trigger}" not found')
+            except ValueError:
+                logger.warning(f'trigger: "{trigger}" more than one similarly-named textual inversion models')
 
         with (
-            ModelPatcher.apply_ti(tokenizer_info.context.model, text_encoder_info.context.model, ti_list) as (
+            ModelPatcher.apply_ti(tokenizer_info.model, text_encoder_info.model, ti_list) as (
                 tokenizer,
                 ti_manager,
             ),
@@ -239,7 +244,7 @@ class SDXLPromptInvocationBase:
             # Apply the LoRA after text_encoder has been moved to its target device for faster patching.
             ModelPatcher.apply_lora(text_encoder, _lora_loader(), lora_prefix),
             # Apply CLIP Skip after LoRA to prevent LoRA application from failing on skipped layers.
-            ModelPatcher.apply_clip_skip(text_encoder_info.context.model, clip_field.skipped_layers),
+            ModelPatcher.apply_clip_skip(text_encoder_info.model, clip_field.skipped_layers),
         ):
             compel = Compel(
                 tokenizer=tokenizer,
@@ -357,6 +362,7 @@ class SDXLCompelPromptInvocation(BaseInvocation, SDXLPromptInvocationBase):
                 dim=1,
             )
 
+        assert c2_pooled is not None
         conditioning_data = ConditioningFieldData(
             conditionings=[
                 SDXLConditioningInfo(
@@ -410,6 +416,7 @@ class SDXLRefinerCompelPromptInvocation(BaseInvocation, SDXLPromptInvocationBase
 
         add_time_ids = torch.tensor([original_size + crop_coords + (self.aesthetic_score,)])
 
+        assert c2_pooled is not None
         conditioning_data = ConditioningFieldData(
             conditionings=[
                 SDXLConditioningInfo(
@@ -459,9 +466,9 @@ class ClipSkipInvocation(BaseInvocation):
 
 
 def get_max_token_count(
-    tokenizer,
+    tokenizer: CLIPTokenizer,
     prompt: Union[FlattenedPrompt, Blend, Conjunction],
-    truncate_if_too_long=False,
+    truncate_if_too_long: bool = False,
 ) -> int:
     if type(prompt) is Blend:
         blend: Blend = prompt
@@ -473,7 +480,9 @@ def get_max_token_count(
         return len(get_tokens_for_prompt_object(tokenizer, prompt, truncate_if_too_long))
 
 
-def get_tokens_for_prompt_object(tokenizer, parsed_prompt: FlattenedPrompt, truncate_if_too_long=True) -> List[str]:
+def get_tokens_for_prompt_object(
+    tokenizer: CLIPTokenizer, parsed_prompt: FlattenedPrompt, truncate_if_too_long: bool = True
+) -> List[str]:
     if type(parsed_prompt) is Blend:
         raise ValueError("Blend is not supported here - you need to get tokens for each of its .children")
 
@@ -486,24 +495,29 @@ def get_tokens_for_prompt_object(tokenizer, parsed_prompt: FlattenedPrompt, trun
         for x in parsed_prompt.children
     ]
     text = " ".join(text_fragments)
-    tokens = tokenizer.tokenize(text)
+    tokens: List[str] = tokenizer.tokenize(text)
     if truncate_if_too_long:
         max_tokens_length = tokenizer.model_max_length - 2  # typically 75
         tokens = tokens[0:max_tokens_length]
     return tokens
 
 
-def log_tokenization_for_conjunction(c: Conjunction, tokenizer, display_label_prefix=None):
+def log_tokenization_for_conjunction(
+    c: Conjunction, tokenizer: CLIPTokenizer, display_label_prefix: Optional[str] = None
+) -> None:
     display_label_prefix = display_label_prefix or ""
     for i, p in enumerate(c.prompts):
         if len(c.prompts) > 1:
             this_display_label_prefix = f"{display_label_prefix}(conjunction part {i + 1}, weight={c.weights[i]})"
         else:
+            assert display_label_prefix is not None
             this_display_label_prefix = display_label_prefix
         log_tokenization_for_prompt_object(p, tokenizer, display_label_prefix=this_display_label_prefix)
 
 
-def log_tokenization_for_prompt_object(p: Union[Blend, FlattenedPrompt], tokenizer, display_label_prefix=None):
+def log_tokenization_for_prompt_object(
+    p: Union[Blend, FlattenedPrompt], tokenizer: CLIPTokenizer, display_label_prefix: Optional[str] = None
+) -> None:
     display_label_prefix = display_label_prefix or ""
     if type(p) is Blend:
         blend: Blend = p
@@ -543,7 +557,12 @@ def log_tokenization_for_prompt_object(p: Union[Blend, FlattenedPrompt], tokeniz
             log_tokenization_for_text(text, tokenizer, display_label=display_label_prefix)
 
 
-def log_tokenization_for_text(text, tokenizer, display_label=None, truncate_if_too_long=False):
+def log_tokenization_for_text(
+    text: str,
+    tokenizer: CLIPTokenizer,
+    display_label: Optional[str] = None,
+    truncate_if_too_long: Optional[bool] = False,
+) -> None:
     """shows how the prompt is tokenized
     # usually tokens have '</w>' to indicate end-of-word,
     # but for readability it has been replaced with ' '

invokeai/app/invocations/controlnet_image_processors.py

@@ -24,13 +24,14 @@ from controlnet_aux import (
 )
 from controlnet_aux.util import HWC3, ade_palette
 from PIL import Image
-from pydantic import BaseModel, ConfigDict, Field, field_validator
+from pydantic import BaseModel, Field, field_validator, model_validator
 
 from invokeai.app.invocations.primitives import ImageField, ImageOutput
+from invokeai.app.invocations.util import validate_begin_end_step, validate_weights
 from invokeai.app.services.image_records.image_records_common import ImageCategory, ResourceOrigin
 from invokeai.app.shared.fields import FieldDescriptions
+from invokeai.backend.image_util.depth_anything import DepthAnythingDetector
 
-from ...backend.model_management import BaseModelType
 from .baseinvocation import (
     BaseInvocation,
     BaseInvocationOutput,
@@ -55,10 +56,7 @@ CONTROLNET_RESIZE_VALUES = Literal[
 class ControlNetModelField(BaseModel):
     """ControlNet model field"""
 
-    model_name: str = Field(description="Name of the ControlNet model")
-    base_model: BaseModelType = Field(description="Base model")
-
-    model_config = ConfigDict(protected_namespaces=())
+    key: str = Field(description="Model config record key for the ControlNet model")
 
 
 class ControlField(BaseModel):
@@ -75,17 +73,16 @@ class ControlField(BaseModel):
     resize_mode: CONTROLNET_RESIZE_VALUES = Field(default="just_resize", description="The resize mode to use")
 
     @field_validator("control_weight")
+    @classmethod
     def validate_control_weight(cls, v):
-        """Validate that all control weights in the valid range"""
-        if isinstance(v, list):
-            for i in v:
-                if i < -1 or i > 2:
-                    raise ValueError("Control weights must be within -1 to 2 range")
-        else:
-            if v < -1 or v > 2:
-                raise ValueError("Control weights must be within -1 to 2 range")
+        validate_weights(v)
         return v
 
+    @model_validator(mode="after")
+    def validate_begin_end_step_percent(self):
+        validate_begin_end_step(self.begin_step_percent, self.end_step_percent)
+        return self
+
 
 @invocation_output("control_output")
 class ControlOutput(BaseInvocationOutput):
@@ -95,17 +92,17 @@ class ControlOutput(BaseInvocationOutput):
     control: ControlField = OutputField(description=FieldDescriptions.control)
 
 
-@invocation("controlnet", title="ControlNet", tags=["controlnet"], category="controlnet", version="1.1.0")
+@invocation("controlnet", title="ControlNet", tags=["controlnet"], category="controlnet", version="1.1.1")
 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_weight: Union[float, List[float]] = InputField(
-        default=1.0, description="The weight given to the ControlNet"
+        default=1.0, ge=-1, le=2, description="The weight given to the ControlNet"
     )
     begin_step_percent: float = InputField(
-        default=0, ge=-1, le=2, description="When the ControlNet is first applied (% of total steps)"
+        default=0, ge=0, le=1, description="When the ControlNet is first applied (% of total steps)"
     )
     end_step_percent: float = InputField(
         default=1, ge=0, le=1, description="When the ControlNet is last applied (% of total steps)"
@@ -113,6 +110,17 @@ class ControlNetInvocation(BaseInvocation):
     control_mode: CONTROLNET_MODE_VALUES = InputField(default="balanced", description="The control mode used")
     resize_mode: CONTROLNET_RESIZE_VALUES = InputField(default="just_resize", description="The resize mode used")
 
+    @field_validator("control_weight")
+    @classmethod
+    def validate_control_weight(cls, v):
+        validate_weights(v)
+        return v
+
+    @model_validator(mode="after")
+    def validate_begin_end_step_percent(self) -> "ControlNetInvocation":
+        validate_begin_end_step(self.begin_step_percent, self.end_step_percent)
+        return self
+
     def invoke(self, context: InvocationContext) -> ControlOutput:
         return ControlOutput(
             control=ControlField(
@@ -591,3 +599,33 @@ class ColorMapImageProcessorInvocation(ImageProcessorInvocation):
         color_map = cv2.resize(color_map, (width, height), interpolation=cv2.INTER_NEAREST)
         color_map = Image.fromarray(color_map)
         return color_map
+
+
+DEPTH_ANYTHING_MODEL_SIZES = Literal["large", "base", "small"]
+
+
+@invocation(
+    "depth_anything_image_processor",
+    title="Depth Anything Processor",
+    tags=["controlnet", "depth", "depth anything"],
+    category="controlnet",
+    version="1.0.0",
+)
+class DepthAnythingImageProcessorInvocation(ImageProcessorInvocation):
+    """Generates a depth map based on the Depth Anything algorithm"""
+
+    model_size: DEPTH_ANYTHING_MODEL_SIZES = InputField(
+        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):
+        depth_anything_detector = DepthAnythingDetector()
+        depth_anything_detector.load_model(model_size=self.model_size)
+
+        if image.mode == "RGBA":
+            image = image.convert("RGB")
+
+        processed_image = depth_anything_detector(image=image, resolution=self.resolution, offload=self.offload)
+        return processed_image

invokeai/app/invocations/ip_adapter.py

@@ -1,8 +1,8 @@
-import os
 from builtins import float
 from typing import List, Union
 
-from pydantic import BaseModel, ConfigDict, Field
+from pydantic import BaseModel, Field, field_validator, model_validator
+from typing_extensions import Self
 
 from invokeai.app.invocations.baseinvocation import (
     BaseInvocation,
@@ -15,23 +15,18 @@ from invokeai.app.invocations.baseinvocation import (
     invocation_output,
 )
 from invokeai.app.invocations.primitives import ImageField
+from invokeai.app.invocations.util import validate_begin_end_step, validate_weights
 from invokeai.app.shared.fields import FieldDescriptions
-from invokeai.backend.model_management.models.base import BaseModelType, ModelType
-from invokeai.backend.model_management.models.ip_adapter import get_ip_adapter_image_encoder_model_id
+from invokeai.backend.model_manager import BaseModelType, ModelType
 
 
+# LS: Consider moving these two classes into model.py
 class IPAdapterModelField(BaseModel):
-    model_name: str = Field(description="Name of the IP-Adapter model")
-    base_model: BaseModelType = Field(description="Base model")
-
-    model_config = ConfigDict(protected_namespaces=())
+    key: str = Field(description="Key to the IP-Adapter model")
 
 
 class CLIPVisionModelField(BaseModel):
-    model_name: str = Field(description="Name of the CLIP Vision image encoder model")
-    base_model: BaseModelType = Field(description="Base model (usually 'Any')")
-
-    model_config = ConfigDict(protected_namespaces=())
+    key: str = Field(description="Key to the CLIP Vision image encoder model")
 
 
 class IPAdapterField(BaseModel):
@@ -39,7 +34,6 @@ class IPAdapterField(BaseModel):
     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.")
     weight: Union[float, List[float]] = Field(default=1, description="The weight given to the ControlNet")
-    # weight: float = Field(default=1.0, ge=0, description="The weight of the IP-Adapter.")
     begin_step_percent: float = Field(
         default=0, ge=0, le=1, description="When the IP-Adapter is first applied (% of total steps)"
     )
@@ -47,6 +41,17 @@ class IPAdapterField(BaseModel):
         default=1, ge=0, le=1, description="When the IP-Adapter is last applied (% of total steps)"
     )
 
+    @field_validator("weight")
+    @classmethod
+    def validate_ip_adapter_weight(cls, v: float) -> float:
+        validate_weights(v)
+        return v
+
+    @model_validator(mode="after")
+    def validate_begin_end_step_percent(self) -> Self:
+        validate_begin_end_step(self.begin_step_percent, self.end_step_percent)
+        return self
+
 
 @invocation_output("ip_adapter_output")
 class IPAdapterOutput(BaseInvocationOutput):
@@ -54,7 +59,7 @@ class IPAdapterOutput(BaseInvocationOutput):
     ip_adapter: IPAdapterField = OutputField(description=FieldDescriptions.ip_adapter, title="IP-Adapter")
 
 
-@invocation("ip_adapter", title="IP-Adapter", tags=["ip_adapter", "control"], category="ip_adapter", version="1.1.0")
+@invocation("ip_adapter", title="IP-Adapter", tags=["ip_adapter", "control"], category="ip_adapter", version="1.1.1")
 class IPAdapterInvocation(BaseInvocation):
     """Collects IP-Adapter info to pass to other nodes."""
 
@@ -64,36 +69,37 @@ class IPAdapterInvocation(BaseInvocation):
         description="The IP-Adapter model.", title="IP-Adapter Model", input=Input.Direct, ui_order=-1
     )
 
-    # weight: float = InputField(default=1.0, description="The weight of the IP-Adapter.", ui_type=UIType.Float)
     weight: Union[float, List[float]] = InputField(
-        default=1, ge=-1, description="The weight given to the IP-Adapter", title="Weight"
+        default=1, description="The weight given to the IP-Adapter", title="Weight"
     )
-
     begin_step_percent: float = InputField(
-        default=0, ge=-1, le=2, description="When the IP-Adapter is first applied (% of total steps)"
+        default=0, ge=0, le=1, description="When the IP-Adapter is first applied (% of total steps)"
     )
     end_step_percent: float = InputField(
         default=1, ge=0, le=1, description="When the IP-Adapter is last applied (% of total steps)"
     )
 
+    @field_validator("weight")
+    @classmethod
+    def validate_ip_adapter_weight(cls, v: float) -> float:
+        validate_weights(v)
+        return v
+
+    @model_validator(mode="after")
+    def validate_begin_end_step_percent(self) -> Self:
+        validate_begin_end_step(self.begin_step_percent, self.end_step_percent)
+        return self
+
     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.services.model_manager.model_info(
-            self.ip_adapter_model.model_name, self.ip_adapter_model.base_model, ModelType.IPAdapter
-        )
-        # HACK(ryand): This is bad for a couple of reasons: 1) we are bypassing the model manager to read the model
-        # directly, and 2) we are reading from disk every time this invocation is called without caching the result.
-        # A better solution would be to store the image encoder model reference in the IP-Adapter model info, but this
-        # is currently messy due to differences between how the model info is generated when installing a model from
-        # disk vs. downloading the model.
-        image_encoder_model_id = get_ip_adapter_image_encoder_model_id(
-            os.path.join(context.services.configuration.get_config().models_path, ip_adapter_info["path"])
-        )
+        ip_adapter_info = context.services.model_manager.store.get_model(self.ip_adapter_model.key)
+        image_encoder_model_id = ip_adapter_info.image_encoder_model_id
         image_encoder_model_name = image_encoder_model_id.split("/")[-1].strip()
-        image_encoder_model = CLIPVisionModelField(
-            model_name=image_encoder_model_name,
-            base_model=BaseModelType.Any,
+        image_encoder_models = context.services.model_manager.store.search_by_attr(
+            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)
         return IPAdapterOutput(
             ip_adapter=IPAdapterField(
                 image=self.image,

invokeai/app/invocations/latent.py

@@ -1,14 +1,17 @@
 # Copyright (c) 2023 Kyle Schouviller (https://github.com/kyle0654)
 
+import math
 from contextlib import ExitStack
 from functools import singledispatchmethod
-from typing import List, Literal, Optional, Union
+from typing import Any, Iterator, List, Literal, Optional, Tuple, Union
 
 import einops
 import numpy as np
+import numpy.typing as npt
 import torch
 import torchvision.transforms as T
 from diffusers import AutoencoderKL, AutoencoderTiny
+from diffusers.configuration_utils import ConfigMixin
 from diffusers.image_processor import VaeImageProcessor
 from diffusers.models.adapter import T2IAdapter
 from diffusers.models.attention_processor import (
@@ -17,8 +20,10 @@ from diffusers.models.attention_processor import (
     LoRAXFormersAttnProcessor,
     XFormersAttnProcessor,
 )
+from diffusers.models.unets.unet_2d_condition import UNet2DConditionModel
 from diffusers.schedulers import DPMSolverSDEScheduler
 from diffusers.schedulers import SchedulerMixin as Scheduler
+from PIL import Image
 from pydantic import field_validator
 from torchvision.transforms.functional import resize as tv_resize
 
@@ -38,13 +43,13 @@ from invokeai.app.shared.fields import FieldDescriptions
 from invokeai.app.util.controlnet_utils import prepare_control_image
 from invokeai.app.util.step_callback import stable_diffusion_step_callback
 from invokeai.backend.ip_adapter.ip_adapter import IPAdapter, IPAdapterPlus
-from invokeai.backend.model_management.models import ModelType, SilenceWarnings
+from invokeai.backend.lora import LoRAModelRaw
+from invokeai.backend.model_manager import BaseModelType, LoadedModel
+from invokeai.backend.model_patcher import ModelPatcher
+from invokeai.backend.stable_diffusion import PipelineIntermediateState, set_seamless
 from invokeai.backend.stable_diffusion.diffusion.conditioning_data import ConditioningData, IPAdapterConditioningInfo
+from invokeai.backend.util.silence_warnings import SilenceWarnings
 
-from ...backend.model_management.lora import ModelPatcher
-from ...backend.model_management.models import BaseModelType
-from ...backend.model_management.seamless import set_seamless
-from ...backend.stable_diffusion import PipelineIntermediateState
 from ...backend.stable_diffusion.diffusers_pipeline import (
     ControlNetData,
     IPAdapterData,
@@ -76,7 +81,9 @@ if choose_torch_device() == torch.device("mps"):
 
 DEFAULT_PRECISION = choose_precision(choose_torch_device())
 
-SAMPLER_NAME_VALUES = Literal[tuple(SCHEDULER_MAP.keys())]
+SAMPLER_NAME_VALUES = Literal[
+    tuple(SCHEDULER_MAP.keys())
+]  # FIXME: "Invalid type alias". This defeats static type checking.
 
 # HACK: Many nodes are currently hard-coded to use a fixed latent scale factor of 8. This is fragile, and will need to
 # be addressed if future models use a different latent scale factor. Also, note that there may be places where the scale
@@ -130,10 +137,10 @@ class CreateDenoiseMaskInvocation(BaseInvocation):
         ui_order=4,
     )
 
-    def prep_mask_tensor(self, mask_image):
+    def prep_mask_tensor(self, mask_image: Image) -> torch.Tensor:
         if mask_image.mode != "L":
             mask_image = mask_image.convert("L")
-        mask_tensor = image_resized_to_grid_as_tensor(mask_image, normalize=False)
+        mask_tensor: torch.Tensor = image_resized_to_grid_as_tensor(mask_image, normalize=False)
         if mask_tensor.dim() == 3:
             mask_tensor = mask_tensor.unsqueeze(0)
         # if shape is not None:
@@ -144,24 +151,24 @@ class CreateDenoiseMaskInvocation(BaseInvocation):
     def invoke(self, context: InvocationContext) -> DenoiseMaskOutput:
         if self.image is not None:
             image = context.services.images.get_pil_image(self.image.image_name)
-            image = image_resized_to_grid_as_tensor(image.convert("RGB"))
-            if image.dim() == 3:
-                image = image.unsqueeze(0)
+            image_tensor = image_resized_to_grid_as_tensor(image.convert("RGB"))
+            if image_tensor.dim() == 3:
+                image_tensor = image_tensor.unsqueeze(0)
         else:
-            image = None
+            image_tensor = None
 
         mask = self.prep_mask_tensor(
             context.services.images.get_pil_image(self.mask.image_name),
         )
 
-        if image is not None:
-            vae_info = context.services.model_manager.get_model(
+        if image_tensor is not None:
+            vae_info = context.services.model_manager.load_model_by_key(
                 **self.vae.vae.model_dump(),
                 context=context,
             )
 
-            img_mask = tv_resize(mask, image.shape[-2:], T.InterpolationMode.BILINEAR, antialias=False)
-            masked_image = image * torch.where(img_mask < 0.5, 0.0, 1.0)
+            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)
             # TODO:
             masked_latents = ImageToLatentsInvocation.vae_encode(vae_info, self.fp32, self.tiled, masked_image.clone())
 
@@ -188,7 +195,7 @@ def get_scheduler(
     seed: int,
 ) -> Scheduler:
     scheduler_class, scheduler_extra_config = SCHEDULER_MAP.get(scheduler_name, SCHEDULER_MAP["ddim"])
-    orig_scheduler_info = context.services.model_manager.get_model(
+    orig_scheduler_info = context.services.model_manager.load_model_by_key(
         **scheduler_info.model_dump(),
         context=context,
     )
@@ -199,7 +206,7 @@ def get_scheduler(
         scheduler_config = scheduler_config["_backup"]
     scheduler_config = {
         **scheduler_config,
-        **scheduler_extra_config,
+        **scheduler_extra_config,  # FIXME
         "_backup": scheduler_config,
     }
 
@@ -212,6 +219,7 @@ def get_scheduler(
     # hack copied over from generate.py
     if not hasattr(scheduler, "uses_inpainting_model"):
         scheduler.uses_inpainting_model = lambda: False
+    assert isinstance(scheduler, Scheduler)
     return scheduler
 
 
@@ -220,7 +228,7 @@ def get_scheduler(
     title="Denoise Latents",
     tags=["latents", "denoise", "txt2img", "t2i", "t2l", "img2img", "i2i", "l2l"],
     category="latents",
-    version="1.5.0",
+    version="1.5.1",
 )
 class DenoiseLatentsInvocation(BaseInvocation):
     """Denoises noisy latents to decodable images"""
@@ -279,7 +287,7 @@ class DenoiseLatentsInvocation(BaseInvocation):
         ui_order=7,
     )
     cfg_rescale_multiplier: float = InputField(
-        default=0, ge=0, lt=1, description=FieldDescriptions.cfg_rescale_multiplier
+        title="CFG Rescale Multiplier", default=0, ge=0, lt=1, description=FieldDescriptions.cfg_rescale_multiplier
     )
     latents: Optional[LatentsField] = InputField(
         default=None,
@@ -295,7 +303,7 @@ class DenoiseLatentsInvocation(BaseInvocation):
     )
 
     @field_validator("cfg_scale")
-    def ge_one(cls, v):
+    def ge_one(cls, v: Union[List[float], float]) -> Union[List[float], float]:
         """validate that all cfg_scale values are >= 1"""
         if isinstance(v, list):
             for i in v:
@@ -325,9 +333,9 @@ class DenoiseLatentsInvocation(BaseInvocation):
     def get_conditioning_data(
         self,
         context: InvocationContext,
-        scheduler,
-        unet,
-        seed,
+        scheduler: Scheduler,
+        unet: UNet2DConditionModel,
+        seed: int,
     ) -> ConditioningData:
         positive_cond_data = context.services.latents.get(self.positive_conditioning.conditioning_name)
         c = positive_cond_data.conditionings[0].to(device=unet.device, dtype=unet.dtype)
@@ -350,7 +358,7 @@ class DenoiseLatentsInvocation(BaseInvocation):
             ),
         )
 
-        conditioning_data = conditioning_data.add_scheduler_args_if_applicable(
+        conditioning_data = conditioning_data.add_scheduler_args_if_applicable(  # FIXME
             scheduler,
             # for ddim scheduler
             eta=0.0,  # ddim_eta
@@ -362,8 +370,8 @@ class DenoiseLatentsInvocation(BaseInvocation):
 
     def create_pipeline(
         self,
-        unet,
-        scheduler,
+        unet: UNet2DConditionModel,
+        scheduler: Scheduler,
     ) -> StableDiffusionGeneratorPipeline:
         # TODO:
         # configure_model_padding(
@@ -374,10 +382,10 @@ class DenoiseLatentsInvocation(BaseInvocation):
 
         class FakeVae:
             class FakeVaeConfig:
-                def __init__(self):
+                def __init__(self) -> None:
                     self.block_out_channels = [0]
 
-            def __init__(self):
+            def __init__(self) -> None:
                 self.config = FakeVae.FakeVaeConfig()
 
         return StableDiffusionGeneratorPipeline(
@@ -394,11 +402,11 @@ class DenoiseLatentsInvocation(BaseInvocation):
     def prep_control_data(
         self,
         context: InvocationContext,
-        control_input: Union[ControlField, List[ControlField]],
+        control_input: Optional[Union[ControlField, List[ControlField]]],
         latents_shape: List[int],
         exit_stack: ExitStack,
         do_classifier_free_guidance: bool = True,
-    ) -> List[ControlNetData]:
+    ) -> Optional[List[ControlNetData]]:
         # Assuming fixed dimensional scaling of LATENT_SCALE_FACTOR.
         control_height_resize = latents_shape[2] * LATENT_SCALE_FACTOR
         control_width_resize = latents_shape[3] * LATENT_SCALE_FACTOR
@@ -421,10 +429,8 @@ class DenoiseLatentsInvocation(BaseInvocation):
         controlnet_data = []
         for control_info in control_list:
             control_model = exit_stack.enter_context(
-                context.services.model_manager.get_model(
-                    model_name=control_info.control_model.model_name,
-                    model_type=ModelType.ControlNet,
-                    base_model=control_info.control_model.base_model,
+                context.services.model_manager.load_model_by_key(
+                    key=control_info.control_model.key,
                     context=context,
                 )
             )
@@ -489,27 +495,25 @@ 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.services.model_manager.get_model(
-                    model_name=single_ip_adapter.ip_adapter_model.model_name,
-                    model_type=ModelType.IPAdapter,
-                    base_model=single_ip_adapter.ip_adapter_model.base_model,
+                context.services.model_manager.load_model_by_key(
+                    key=single_ip_adapter.ip_adapter_model.key,
                     context=context,
                 )
             )
 
-            image_encoder_model_info = context.services.model_manager.get_model(
-                model_name=single_ip_adapter.image_encoder_model.model_name,
-                model_type=ModelType.CLIPVision,
-                base_model=single_ip_adapter.image_encoder_model.base_model,
+            image_encoder_model_info = context.services.model_manager.load_model_by_key(
+                key=single_ip_adapter.image_encoder_model.key,
                 context=context,
             )
 
             # `single_ip_adapter.image` could be a list or a single ImageField. Normalize to a list here.
-            single_ipa_images = single_ip_adapter.image
-            if not isinstance(single_ipa_images, list):
-                single_ipa_images = [single_ipa_images]
+            single_ipa_image_fields = single_ip_adapter.image
+            if not isinstance(single_ipa_image_fields, list):
+                single_ipa_image_fields = [single_ipa_image_fields]
 
-            single_ipa_images = [context.services.images.get_pil_image(image.image_name) for image in single_ipa_images]
+            single_ipa_images = [
+                context.services.images.get_pil_image(image.image_name) for image in single_ipa_image_fields
+            ]
 
             # 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.
@@ -553,23 +557,19 @@ class DenoiseLatentsInvocation(BaseInvocation):
 
         t2i_adapter_data = []
         for t2i_adapter_field in t2i_adapter:
-            t2i_adapter_model_info = context.services.model_manager.get_model(
-                model_name=t2i_adapter_field.t2i_adapter_model.model_name,
-                model_type=ModelType.T2IAdapter,
-                base_model=t2i_adapter_field.t2i_adapter_model.base_model,
+            t2i_adapter_model_info = context.services.model_manager.load_model_by_key(
+                key=t2i_adapter_field.t2i_adapter_model.key,
                 context=context,
             )
             image = context.services.images.get_pil_image(t2i_adapter_field.image.image_name)
 
             # The max_unet_downscale is the maximum amount that the UNet model downscales the latent image internally.
-            if t2i_adapter_field.t2i_adapter_model.base_model == BaseModelType.StableDiffusion1:
+            if t2i_adapter_model_info.base == BaseModelType.StableDiffusion1:
                 max_unet_downscale = 8
-            elif t2i_adapter_field.t2i_adapter_model.base_model == BaseModelType.StableDiffusionXL:
+            elif t2i_adapter_model_info.base == BaseModelType.StableDiffusionXL:
                 max_unet_downscale = 4
             else:
-                raise ValueError(
-                    f"Unexpected T2I-Adapter base model type: '{t2i_adapter_field.t2i_adapter_model.base_model}'."
-                )
+                raise ValueError(f"Unexpected T2I-Adapter base model type: '{t2i_adapter_model_info.base}'.")
 
             t2i_adapter_model: T2IAdapter
             with t2i_adapter_model_info as t2i_adapter_model:
@@ -592,7 +592,7 @@ class DenoiseLatentsInvocation(BaseInvocation):
                     do_classifier_free_guidance=False,
                     width=t2i_input_width,
                     height=t2i_input_height,
-                    num_channels=t2i_adapter_model.config.in_channels,
+                    num_channels=t2i_adapter_model.config["in_channels"],  # mypy treats this as a FrozenDict
                     device=t2i_adapter_model.device,
                     dtype=t2i_adapter_model.dtype,
                     resize_mode=t2i_adapter_field.resize_mode,
@@ -617,7 +617,15 @@ class DenoiseLatentsInvocation(BaseInvocation):
 
     # original idea by https://github.com/AmericanPresidentJimmyCarter
     # TODO: research more for second order schedulers timesteps
-    def init_scheduler(self, scheduler, device, steps, denoising_start, denoising_end):
+    def init_scheduler(
+        self,
+        scheduler: Union[Scheduler, ConfigMixin],
+        device: torch.device,
+        steps: int,
+        denoising_start: float,
+        denoising_end: float,
+    ) -> Tuple[int, List[int], int]:
+        assert isinstance(scheduler, ConfigMixin)
         if scheduler.config.get("cpu_only", False):
             scheduler.set_timesteps(steps, device="cpu")
             timesteps = scheduler.timesteps.to(device=device)
@@ -629,11 +637,11 @@ class DenoiseLatentsInvocation(BaseInvocation):
         _timesteps = timesteps[:: scheduler.order]
 
         # get start timestep index
-        t_start_val = int(round(scheduler.config.num_train_timesteps * (1 - denoising_start)))
+        t_start_val = int(round(scheduler.config["num_train_timesteps"] * (1 - denoising_start)))
         t_start_idx = len(list(filter(lambda ts: ts >= t_start_val, _timesteps)))
 
         # get end timestep index
-        t_end_val = int(round(scheduler.config.num_train_timesteps * (1 - denoising_end)))
+        t_end_val = int(round(scheduler.config["num_train_timesteps"] * (1 - denoising_end)))
         t_end_idx = len(list(filter(lambda ts: ts >= t_end_val, _timesteps[t_start_idx:])))
 
         # apply order to indexes
@@ -646,7 +654,9 @@ class DenoiseLatentsInvocation(BaseInvocation):
 
         return num_inference_steps, timesteps, init_timestep
 
-    def prep_inpaint_mask(self, context, latents):
+    def prep_inpaint_mask(
+        self, context: InvocationContext, latents: torch.Tensor
+    ) -> Tuple[Optional[torch.Tensor], Optional[torch.Tensor]]:
         if self.denoise_mask is None:
             return None, None
 
@@ -699,31 +709,36 @@ class DenoiseLatentsInvocation(BaseInvocation):
             graph_execution_state = context.services.graph_execution_manager.get(context.graph_execution_state_id)
             source_node_id = graph_execution_state.prepared_source_mapping[self.id]
 
-            def step_callback(state: PipelineIntermediateState):
-                self.dispatch_progress(context, source_node_id, state, self.unet.unet.base_model)
+            # get the unet's config so that we can pass the base to dispatch_progress()
+            unet_config = context.services.model_manager.store.get_model(self.unet.unet.key)
 
-            def _lora_loader():
+            def step_callback(state: PipelineIntermediateState) -> None:
+                self.dispatch_progress(context, source_node_id, state, unet_config.base)
+
+            def _lora_loader() -> Iterator[Tuple[LoRAModelRaw, float]]:
                 for lora in self.unet.loras:
-                    lora_info = context.services.model_manager.get_model(
+                    lora_info = context.services.model_manager.load_model_by_key(
                         **lora.model_dump(exclude={"weight"}),
                         context=context,
                     )
-                    yield (lora_info.context.model, lora.weight)
+                    yield (lora_info.model, lora.weight)
                     del lora_info
                 return
 
-            unet_info = context.services.model_manager.get_model(
+            unet_info = context.services.model_manager.load_model_by_key(
                 **self.unet.unet.model_dump(),
                 context=context,
             )
+            assert isinstance(unet_info.model, UNet2DConditionModel)
             with (
                 ExitStack() as exit_stack,
-                ModelPatcher.apply_freeu(unet_info.context.model, self.unet.freeu_config),
-                set_seamless(unet_info.context.model, self.unet.seamless_axes),
+                ModelPatcher.apply_freeu(unet_info.model, self.unet.freeu_config),
+                set_seamless(unet_info.model, self.unet.seamless_axes),  # FIXME
                 unet_info as unet,
                 # Apply the LoRA after unet has been moved to its target device for faster patching.
                 ModelPatcher.apply_lora_unet(unet, _lora_loader()),
             ):
+                assert isinstance(unet, UNet2DConditionModel)
                 latents = latents.to(device=unet.device, dtype=unet.dtype)
                 if noise is not None:
                     noise = noise.to(device=unet.device, dtype=unet.dtype)
@@ -821,12 +836,13 @@ class LatentsToImageInvocation(BaseInvocation, WithMetadata):
     def invoke(self, context: InvocationContext) -> ImageOutput:
         latents = context.services.latents.get(self.latents.latents_name)
 
-        vae_info = context.services.model_manager.get_model(
+        vae_info = context.services.model_manager.load_model_by_key(
             **self.vae.vae.model_dump(),
             context=context,
         )
 
-        with set_seamless(vae_info.context.model, self.vae.seamless_axes), vae_info as vae:
+        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)
@@ -1015,8 +1031,9 @@ class ImageToLatentsInvocation(BaseInvocation):
     fp32: bool = InputField(default=DEFAULT_PRECISION == "float32", description=FieldDescriptions.fp32)
 
     @staticmethod
-    def vae_encode(vae_info, upcast, tiled, image_tensor):
+    def vae_encode(vae_info: LoadedModel, upcast: bool, tiled: bool, image_tensor: torch.Tensor) -> torch.Tensor:
         with vae_info as vae:
+            assert isinstance(vae, torch.nn.Module)
             orig_dtype = vae.dtype
             if upcast:
                 vae.to(dtype=torch.float32)
@@ -1062,7 +1079,7 @@ class ImageToLatentsInvocation(BaseInvocation):
     def invoke(self, context: InvocationContext) -> LatentsOutput:
         image = context.services.images.get_pil_image(self.image.image_name)
 
-        vae_info = context.services.model_manager.get_model(
+        vae_info = context.services.model_manager.load_model_by_key(
             **self.vae.vae.model_dump(),
             context=context,
         )
@@ -1081,14 +1098,19 @@ class ImageToLatentsInvocation(BaseInvocation):
     @singledispatchmethod
     @staticmethod
     def _encode_to_tensor(vae: AutoencoderKL, image_tensor: torch.FloatTensor) -> torch.FloatTensor:
+        assert isinstance(vae, torch.nn.Module)
         image_tensor_dist = vae.encode(image_tensor).latent_dist
-        latents = image_tensor_dist.sample().to(dtype=vae.dtype)  # FIXME: uses torch.randn. make reproducible!
+        latents: torch.Tensor = image_tensor_dist.sample().to(
+            dtype=vae.dtype
+        )  # FIXME: uses torch.randn. make reproducible!
         return latents
 
     @_encode_to_tensor.register
     @staticmethod
     def _(vae: AutoencoderTiny, image_tensor: torch.FloatTensor) -> torch.FloatTensor:
-        return vae.encode(image_tensor).latents
+        assert isinstance(vae, torch.nn.Module)
+        latents: torch.FloatTensor = vae.encode(image_tensor).latents
+        return latents
 
 
 @invocation(
@@ -1121,7 +1143,12 @@ class BlendLatentsInvocation(BaseInvocation):
         # TODO:
         device = choose_torch_device()
 
-        def slerp(t, v0, v1, DOT_THRESHOLD=0.9995):
+        def slerp(
+            t: Union[float, npt.NDArray[Any]],  # FIXME: maybe use np.float32 here?
+            v0: Union[torch.Tensor, npt.NDArray[Any]],
+            v1: Union[torch.Tensor, npt.NDArray[Any]],
+            DOT_THRESHOLD: float = 0.9995,
+        ) -> Union[torch.Tensor, npt.NDArray[Any]]:
             """
             Spherical linear interpolation
             Args:
@@ -1154,12 +1181,16 @@ class BlendLatentsInvocation(BaseInvocation):
                 v2 = s0 * v0 + s1 * v1
 
             if inputs_are_torch:
-                v2 = torch.from_numpy(v2).to(device)
-
-            return v2
+                v2_torch: torch.Tensor = torch.from_numpy(v2).to(device)
+                return v2_torch
+            else:
+                assert isinstance(v2, np.ndarray)
+                return v2
 
         # blend
-        blended_latents = slerp(self.alpha, latents_a, latents_b)
+        bl = slerp(self.alpha, latents_a, latents_b)
+        assert isinstance(bl, torch.Tensor)
+        blended_latents: torch.Tensor = bl  # for type checking convenience
 
         # https://discuss.huggingface.co/t/memory-usage-by-later-pipeline-stages/23699
         blended_latents = blended_latents.to("cpu")
@@ -1228,3 +1259,61 @@ class CropLatentsCoreInvocation(BaseInvocation):
         context.services.latents.save(name, cropped_latents)
 
         return build_latents_output(latents_name=name, latents=cropped_latents)
+
+
+@invocation_output("ideal_size_output")
+class IdealSizeOutput(BaseInvocationOutput):
+    """Base class for invocations that output an image"""
+
+    width: int = OutputField(description="The ideal width of the image (in pixels)")
+    height: int = OutputField(description="The ideal height of the image (in pixels)")
+
+
+@invocation(
+    "ideal_size",
+    title="Ideal Size",
+    tags=["latents", "math", "ideal_size"],
+    version="1.0.2",
+)
+class IdealSizeInvocation(BaseInvocation):
+    """Calculates the ideal size for generation to avoid duplication"""
+
+    width: int = InputField(default=1024, description="Final image width")
+    height: int = InputField(default=576, description="Final image height")
+    unet: UNetField = InputField(default=None, description=FieldDescriptions.unet)
+    multiplier: float = InputField(
+        default=1.0,
+        description="Amount to multiply the model's dimensions by when calculating the ideal size (may result in initial generation artifacts if too large)",
+    )
+
+    def trim_to_multiple_of(self, *args: int, multiple_of: int = LATENT_SCALE_FACTOR) -> Tuple[int, ...]:
+        return tuple((x - x % multiple_of) for x in args)
+
+    def invoke(self, context: InvocationContext) -> IdealSizeOutput:
+        unet_config = context.services.model_manager.load_model_by_key(
+            **self.unet.unet.model_dump(),
+            context=context,
+        )
+        aspect = self.width / self.height
+        dimension: float = 512
+        if unet_config.base == BaseModelType.StableDiffusion2:
+            dimension = 768
+        elif unet_config.base == BaseModelType.StableDiffusionXL:
+            dimension = 1024
+        dimension = dimension * self.multiplier
+        min_dimension = math.floor(dimension * 0.5)
+        model_area = dimension * dimension  # hardcoded for now since all models are trained on square images
+
+        if aspect > 1.0:
+            init_height = max(min_dimension, math.sqrt(model_area / aspect))
+            init_width = init_height * aspect
+        else:
+            init_width = max(min_dimension, math.sqrt(model_area * aspect))
+            init_height = init_width / aspect
+
+        scaled_width, scaled_height = self.trim_to_multiple_of(
+            math.floor(init_width),
+            math.floor(init_height),
+        )
+
+        return IdealSizeOutput(width=scaled_width, height=scaled_height)

invokeai/app/invocations/model.py

@@ -1,12 +1,12 @@
 import copy
 from typing import List, Optional
 
-from pydantic import BaseModel, ConfigDict, Field
+from pydantic import BaseModel, Field
 
 from invokeai.app.shared.fields import FieldDescriptions
 from invokeai.app.shared.models import FreeUConfig
 
-from ...backend.model_management import BaseModelType, ModelType, SubModelType
+from ...backend.model_manager import SubModelType
 from .baseinvocation import (
     BaseInvocation,
     BaseInvocationOutput,
@@ -20,12 +20,8 @@ from .baseinvocation import (
 
 
 class ModelInfo(BaseModel):
-    model_name: str = Field(description="Info to load submodel")
-    base_model: BaseModelType = Field(description="Base model")
-    model_type: ModelType = Field(description="Info to load submodel")
-    submodel: Optional[SubModelType] = Field(default=None, description="Info to load submodel")
-
-    model_config = ConfigDict(protected_namespaces=())
+    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 LoraInfo(ModelInfo):
@@ -55,7 +51,7 @@ class VaeField(BaseModel):
 
 @invocation_output("unet_output")
 class UNetOutput(BaseInvocationOutput):
-    """Base class for invocations that output a UNet field"""
+    """Base class for invocations that output a UNet field."""
 
     unet: UNetField = OutputField(description=FieldDescriptions.unet, title="UNet")
 
@@ -84,20 +80,13 @@ class ModelLoaderOutput(UNetOutput, CLIPOutput, VAEOutput):
 class MainModelField(BaseModel):
     """Main model field"""
 
-    model_name: str = Field(description="Name of the model")
-    base_model: BaseModelType = Field(description="Base model")
-    model_type: ModelType = Field(description="Model Type")
-
-    model_config = ConfigDict(protected_namespaces=())
+    key: str = Field(description="Model key")
 
 
 class LoRAModelField(BaseModel):
     """LoRA model field"""
 
-    model_name: str = Field(description="Name of the LoRA model")
-    base_model: BaseModelType = Field(description="Base model")
-
-    model_config = ConfigDict(protected_namespaces=())
+    key: str = Field(description="LoRA model key")
 
 
 @invocation(
@@ -114,85 +103,40 @@ class MainModelLoaderInvocation(BaseInvocation):
     # TODO: precision?
 
     def invoke(self, context: InvocationContext) -> ModelLoaderOutput:
-        base_model = self.model.base_model
-        model_name = self.model.model_name
-        model_type = ModelType.Main
+        key = self.model.key
 
         # TODO: not found exceptions
-        if not context.services.model_manager.model_exists(
-            model_name=model_name,
-            base_model=base_model,
-            model_type=model_type,
-        ):
-            raise Exception(f"Unknown {base_model} {model_type} model: {model_name}")
-
-        """
-        if not context.services.model_manager.model_exists(
-            model_name=self.model_name,
-            model_type=SDModelType.Diffusers,
-            submodel=SDModelType.Tokenizer,
-        ):
-            raise Exception(
-                f"Failed to find tokenizer submodel in {self.model_name}! Check if model corrupted"
-            )
-
-        if not context.services.model_manager.model_exists(
-            model_name=self.model_name,
-            model_type=SDModelType.Diffusers,
-            submodel=SDModelType.TextEncoder,
-        ):
-            raise Exception(
-                f"Failed to find text_encoder submodel in {self.model_name}! Check if model corrupted"
-            )
-
-        if not context.services.model_manager.model_exists(
-            model_name=self.model_name,
-            model_type=SDModelType.Diffusers,
-            submodel=SDModelType.UNet,
-        ):
-            raise Exception(
-                f"Failed to find unet submodel from {self.model_name}! Check if model corrupted"
-            )
-        """
+        if not context.services.model_manager.store.exists(key):
+            raise Exception(f"Unknown model {key}")
 
         return ModelLoaderOutput(
             unet=UNetField(
                 unet=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.UNet,
+                    key=key,
+                    submodel_type=SubModelType.UNet,
                 ),
                 scheduler=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.Scheduler,
+                    key=key,
+                    submodel_type=SubModelType.Scheduler,
                 ),
                 loras=[],
             ),
             clip=ClipField(
                 tokenizer=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.Tokenizer,
+                    key=key,
+                    submodel_type=SubModelType.Tokenizer,
                 ),
                 text_encoder=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.TextEncoder,
+                    key=key,
+                    submodel_type=SubModelType.TextEncoder,
                 ),
                 loras=[],
                 skipped_layers=0,
             ),
             vae=VaeField(
                 vae=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.Vae,
+                    key=key,
+                    submodel_type=SubModelType.Vae,
                 ),
             ),
         )
@@ -229,21 +173,16 @@ class LoraLoaderInvocation(BaseInvocation):
         if self.lora is None:
             raise Exception("No LoRA provided")
 
-        base_model = self.lora.base_model
-        lora_name = self.lora.model_name
+        lora_key = self.lora.key
 
-        if not context.services.model_manager.model_exists(
-            base_model=base_model,
-            model_name=lora_name,
-            model_type=ModelType.Lora,
-        ):
-            raise Exception(f"Unkown lora name: {lora_name}!")
+        if not context.services.model_manager.store.exists(lora_key):
+            raise Exception(f"Unkown lora: {lora_key}!")
 
-        if self.unet is not None and any(lora.model_name == lora_name for lora in self.unet.loras):
-            raise Exception(f'Lora "{lora_name}" already applied to unet')
+        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.clip is not None and any(lora.model_name == lora_name for lora in self.clip.loras):
-            raise Exception(f'Lora "{lora_name}" already applied to clip')
+        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')
 
         output = LoraLoaderOutput()
 
@@ -251,10 +190,8 @@ class LoraLoaderInvocation(BaseInvocation):
             output.unet = copy.deepcopy(self.unet)
             output.unet.loras.append(
                 LoraInfo(
-                    base_model=base_model,
-                    model_name=lora_name,
-                    model_type=ModelType.Lora,
-                    submodel=None,
+                    key=lora_key,
+                    submodel_type=None,
                     weight=self.weight,
                 )
             )
@@ -263,10 +200,8 @@ class LoraLoaderInvocation(BaseInvocation):
             output.clip = copy.deepcopy(self.clip)
             output.clip.loras.append(
                 LoraInfo(
-                    base_model=base_model,
-                    model_name=lora_name,
-                    model_type=ModelType.Lora,
-                    submodel=None,
+                    key=lora_key,
+                    submodel_type=None,
                     weight=self.weight,
                 )
             )
@@ -318,24 +253,19 @@ class SDXLLoraLoaderInvocation(BaseInvocation):
         if self.lora is None:
             raise Exception("No LoRA provided")
 
-        base_model = self.lora.base_model
-        lora_name = self.lora.model_name
+        lora_key = self.lora.key
 
-        if not context.services.model_manager.model_exists(
-            base_model=base_model,
-            model_name=lora_name,
-            model_type=ModelType.Lora,
-        ):
-            raise Exception(f"Unknown lora name: {lora_name}!")
+        if not context.services.model_manager.store.exists(lora_key):
+            raise Exception(f"Unknown lora: {lora_key}!")
 
-        if self.unet is not None and any(lora.model_name == lora_name for lora in self.unet.loras):
-            raise Exception(f'Lora "{lora_name}" already applied to unet')
+        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.clip is not None and any(lora.model_name == lora_name for lora in self.clip.loras):
-            raise Exception(f'Lora "{lora_name}" already applied to clip')
+        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.clip2 is not None and any(lora.model_name == lora_name for lora in self.clip2.loras):
-            raise Exception(f'Lora "{lora_name}" already applied to clip2')
+        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')
 
         output = SDXLLoraLoaderOutput()
 
@@ -343,10 +273,8 @@ class SDXLLoraLoaderInvocation(BaseInvocation):
             output.unet = copy.deepcopy(self.unet)
             output.unet.loras.append(
                 LoraInfo(
-                    base_model=base_model,
-                    model_name=lora_name,
-                    model_type=ModelType.Lora,
-                    submodel=None,
+                    key=lora_key,
+                    submodel_type=None,
                     weight=self.weight,
                 )
             )
@@ -355,10 +283,8 @@ class SDXLLoraLoaderInvocation(BaseInvocation):
             output.clip = copy.deepcopy(self.clip)
             output.clip.loras.append(
                 LoraInfo(
-                    base_model=base_model,
-                    model_name=lora_name,
-                    model_type=ModelType.Lora,
-                    submodel=None,
+                    key=lora_key,
+                    submodel_type=None,
                     weight=self.weight,
                 )
             )
@@ -367,10 +293,8 @@ class SDXLLoraLoaderInvocation(BaseInvocation):
             output.clip2 = copy.deepcopy(self.clip2)
             output.clip2.loras.append(
                 LoraInfo(
-                    base_model=base_model,
-                    model_name=lora_name,
-                    model_type=ModelType.Lora,
-                    submodel=None,
+                    key=lora_key,
+                    submodel_type=None,
                     weight=self.weight,
                 )
             )
@@ -381,10 +305,7 @@ class SDXLLoraLoaderInvocation(BaseInvocation):
 class VAEModelField(BaseModel):
     """Vae model field"""
 
-    model_name: str = Field(description="Name of the model")
-    base_model: BaseModelType = Field(description="Base model")
-
-    model_config = ConfigDict(protected_namespaces=())
+    key: str = Field(description="Model's key")
 
 
 @invocation("vae_loader", title="VAE", tags=["vae", "model"], category="model", version="1.0.0")
@@ -398,25 +319,12 @@ class VaeLoaderInvocation(BaseInvocation):
     )
 
     def invoke(self, context: InvocationContext) -> VAEOutput:
-        base_model = self.vae_model.base_model
-        model_name = self.vae_model.model_name
-        model_type = ModelType.Vae
+        key = self.vae_model.key
 
-        if not context.services.model_manager.model_exists(
-            base_model=base_model,
-            model_name=model_name,
-            model_type=model_type,
-        ):
-            raise Exception(f"Unkown vae name: {model_name}!")
-        return VAEOutput(
-            vae=VaeField(
-                vae=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                )
-            )
-        )
+        if not context.services.model_manager.store.exists(key):
+            raise Exception(f"Unkown vae: {key}!")
+
+        return VAEOutput(vae=VaeField(vae=ModelInfo(key=key)))
 
 
 @invocation_output("seamless_output")

invokeai/app/invocations/onnx.py

@@ -1,7 +1,6 @@
 # Copyright (c) 2023 Borisov Sergey (https://github.com/StAlKeR7779)
 
 import inspect
-import re
 
 # from contextlib import ExitStack
 from typing import List, Literal, Union
@@ -9,18 +8,19 @@ from typing import List, Literal, Union
 import numpy as np
 import torch
 from diffusers.image_processor import VaeImageProcessor
-from pydantic import BaseModel, ConfigDict, Field, field_validator
+from pydantic import BaseModel, Field, field_validator
 from tqdm import tqdm
 
 from invokeai.app.invocations.primitives import ConditioningField, ConditioningOutput, ImageField, ImageOutput
 from invokeai.app.services.image_records.image_records_common import ImageCategory, ResourceOrigin
 from invokeai.app.shared.fields import FieldDescriptions
 from invokeai.app.util.step_callback import stable_diffusion_step_callback
-from invokeai.backend import BaseModelType, ModelType, SubModelType
+from invokeai.backend.model_manager import ModelType, SubModelType
+from invokeai.backend.model_patcher import ONNXModelPatcher
 
-from ...backend.model_management import ONNXModelPatcher
 from ...backend.stable_diffusion import PipelineIntermediateState
 from ...backend.util import choose_torch_device
+from ..util.ti_utils import extract_ti_triggers_from_prompt
 from .baseinvocation import (
     BaseInvocation,
     BaseInvocationOutput,
@@ -62,33 +62,33 @@ class ONNXPromptInvocation(BaseInvocation):
     clip: ClipField = InputField(description=FieldDescriptions.clip, input=Input.Connection)
 
     def invoke(self, context: InvocationContext) -> ConditioningOutput:
-        tokenizer_info = context.services.model_manager.get_model(
+        tokenizer_info = context.services.model_manager.load_model_by_key(
             **self.clip.tokenizer.model_dump(),
         )
-        text_encoder_info = context.services.model_manager.get_model(
+        text_encoder_info = context.services.model_manager.load_model_by_key(
             **self.clip.text_encoder.model_dump(),
         )
         with tokenizer_info as orig_tokenizer, text_encoder_info as text_encoder:  # , ExitStack() as stack:
             loras = [
                 (
-                    context.services.model_manager.get_model(**lora.model_dump(exclude={"weight"})).context.model,
+                    context.services.model_manager.load_model_by_key(**lora.model_dump(exclude={"weight"})).model,
                     lora.weight,
                 )
                 for lora in self.clip.loras
             ]
 
             ti_list = []
-            for trigger in re.findall(r"<[a-zA-Z0-9., _-]+>", self.prompt):
+            for trigger in extract_ti_triggers_from_prompt(self.prompt):
                 name = trigger[1:-1]
                 try:
                     ti_list.append(
                         (
                             name,
-                            context.services.model_manager.get_model(
+                            context.services.model_manager.load_model_by_attr(
                                 model_name=name,
-                                base_model=self.clip.text_encoder.base_model,
+                                base_model=text_encoder_info.config.base,
                                 model_type=ModelType.TextualInversion,
-                            ).context.model,
+                            ).model,
                         )
                     )
                 except Exception:
@@ -257,13 +257,13 @@ class ONNXTextToLatentsInvocation(BaseInvocation):
                 eta=0.0,
             )
 
-        unet_info = context.services.model_manager.get_model(**self.unet.unet.model_dump())
+        unet_info = context.services.model_manager.load_model_by_key(**self.unet.unet.model_dump())
 
         with unet_info as unet:  # , ExitStack() as stack:
             # loras = [(stack.enter_context(context.services.model_manager.get_model(**lora.dict(exclude={"weight"}))), lora.weight) for lora in self.unet.loras]
             loras = [
                 (
-                    context.services.model_manager.get_model(**lora.model_dump(exclude={"weight"})).context.model,
+                    context.services.model_manager.load_model_by_key(**lora.model_dump(exclude={"weight"})).model,
                     lora.weight,
                 )
                 for lora in self.unet.loras
@@ -344,9 +344,9 @@ class ONNXLatentsToImageInvocation(BaseInvocation, WithMetadata):
         latents = context.services.latents.get(self.latents.latents_name)
 
         if self.vae.vae.submodel != SubModelType.VaeDecoder:
-            raise Exception(f"Expected vae_decoder, found: {self.vae.vae.model_type}")
+            raise Exception(f"Expected vae_decoder, found: {self.vae.vae.submodel}")
 
-        vae_info = context.services.model_manager.get_model(
+        vae_info = context.services.model_manager.load_model_by_key(
             **self.vae.vae.model_dump(),
         )
 
@@ -400,11 +400,7 @@ class ONNXModelLoaderOutput(BaseInvocationOutput):
 class OnnxModelField(BaseModel):
     """Onnx model field"""
 
-    model_name: str = Field(description="Name of the model")
-    base_model: BaseModelType = Field(description="Base model")
-    model_type: ModelType = Field(description="Model Type")
-
-    model_config = ConfigDict(protected_namespaces=())
+    key: str = Field(description="Model ID")
 
 
 @invocation("onnx_model_loader", title="ONNX Main Model", tags=["onnx", "model"], category="model", version="1.0.0")
@@ -416,93 +412,46 @@ class OnnxModelLoaderInvocation(BaseInvocation):
     )
 
     def invoke(self, context: InvocationContext) -> ONNXModelLoaderOutput:
-        base_model = self.model.base_model
-        model_name = self.model.model_name
-        model_type = ModelType.ONNX
+        model_key = self.model.key
 
         # TODO: not found exceptions
-        if not context.services.model_manager.model_exists(
-            model_name=model_name,
-            base_model=base_model,
-            model_type=model_type,
-        ):
-            raise Exception(f"Unknown {base_model} {model_type} model: {model_name}")
-
-        """
-        if not context.services.model_manager.model_exists(
-            model_name=self.model_name,
-            model_type=SDModelType.Diffusers,
-            submodel=SDModelType.Tokenizer,
-        ):
-            raise Exception(
-                f"Failed to find tokenizer submodel in {self.model_name}! Check if model corrupted"
-            )
-
-        if not context.services.model_manager.model_exists(
-            model_name=self.model_name,
-            model_type=SDModelType.Diffusers,
-            submodel=SDModelType.TextEncoder,
-        ):
-            raise Exception(
-                f"Failed to find text_encoder submodel in {self.model_name}! Check if model corrupted"
-            )
-
-        if not context.services.model_manager.model_exists(
-            model_name=self.model_name,
-            model_type=SDModelType.Diffusers,
-            submodel=SDModelType.UNet,
-        ):
-            raise Exception(
-                f"Failed to find unet submodel from {self.model_name}! Check if model corrupted"
-            )
-        """
+        if not context.services.model_manager.store.exists(model_key):
+            raise Exception(f"Unknown model: {model_key}")
 
         return ONNXModelLoaderOutput(
             unet=UNetField(
                 unet=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.UNet,
+                    key=model_key,
+                    submodel_type=SubModelType.UNet,
                 ),
                 scheduler=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.Scheduler,
+                    key=model_key,
+                    submodel_type=SubModelType.Scheduler,
                 ),
                 loras=[],
             ),
             clip=ClipField(
                 tokenizer=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.Tokenizer,
+                    key=model_key,
+                    submodel_type=SubModelType.Tokenizer,
                 ),
                 text_encoder=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.TextEncoder,
+                    key=model_key,
+                    submodel_type=SubModelType.TextEncoder,
                 ),
                 loras=[],
                 skipped_layers=0,
             ),
             vae_decoder=VaeField(
                 vae=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.VaeDecoder,
+                    key=model_key,
+                    submodel_type=SubModelType.VaeDecoder,
                 ),
             ),
             vae_encoder=VaeField(
                 vae=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.VaeEncoder,
+                    key=model_key,
+                    submodel_type=SubModelType.VaeEncoder,
                 ),
             ),
         )

invokeai/app/invocations/primitives.py

@@ -368,7 +368,7 @@ class LatentsCollectionInvocation(BaseInvocation):
         return LatentsCollectionOutput(collection=self.collection)
 
 
-def build_latents_output(latents_name: str, latents: torch.Tensor, seed: Optional[int] = None):
+def build_latents_output(latents_name: str, latents: torch.Tensor, seed: Optional[int] = None) -> LatentsOutput:
     return LatentsOutput(
         latents=LatentsField(latents_name=latents_name, seed=seed),
         width=latents.size()[3] * 8,

invokeai/app/invocations/sdxl.py

@@ -1,6 +1,6 @@
 from invokeai.app.shared.fields import FieldDescriptions
+from invokeai.backend.model_manager import SubModelType
 
-from ...backend.model_management import ModelType, SubModelType
 from .baseinvocation import (
     BaseInvocation,
     BaseInvocationOutput,
@@ -44,72 +44,52 @@ class SDXLModelLoaderInvocation(BaseInvocation):
     # TODO: precision?
 
     def invoke(self, context: InvocationContext) -> SDXLModelLoaderOutput:
-        base_model = self.model.base_model
-        model_name = self.model.model_name
-        model_type = ModelType.Main
+        model_key = self.model.key
 
         # TODO: not found exceptions
-        if not context.services.model_manager.model_exists(
-            model_name=model_name,
-            base_model=base_model,
-            model_type=model_type,
-        ):
-            raise Exception(f"Unknown {base_model} {model_type} model: {model_name}")
+        if not context.services.model_manager.store.exists(model_key):
+            raise Exception(f"Unknown model: {model_key}")
 
         return SDXLModelLoaderOutput(
             unet=UNetField(
                 unet=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.UNet,
+                    key=model_key,
+                    submodel_type=SubModelType.UNet,
                 ),
                 scheduler=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.Scheduler,
+                    key=model_key,
+                    submodel_type=SubModelType.Scheduler,
                 ),
                 loras=[],
             ),
             clip=ClipField(
                 tokenizer=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.Tokenizer,
+                    key=model_key,
+                    submodel_type=SubModelType.Tokenizer,
                 ),
                 text_encoder=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.TextEncoder,
+                    key=model_key,
+                    submodel_type=SubModelType.TextEncoder,
                 ),
                 loras=[],
                 skipped_layers=0,
             ),
             clip2=ClipField(
                 tokenizer=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.Tokenizer2,
+                    key=model_key,
+                    submodel_type=SubModelType.Tokenizer2,
                 ),
                 text_encoder=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.TextEncoder2,
+                    key=model_key,
+                    submodel_type=SubModelType.TextEncoder2,
                 ),
                 loras=[],
                 skipped_layers=0,
             ),
             vae=VaeField(
                 vae=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.Vae,
+                    key=model_key,
+                    submodel_type=SubModelType.Vae,
                 ),
             ),
         )
@@ -133,56 +113,40 @@ class SDXLRefinerModelLoaderInvocation(BaseInvocation):
     # TODO: precision?
 
     def invoke(self, context: InvocationContext) -> SDXLRefinerModelLoaderOutput:
-        base_model = self.model.base_model
-        model_name = self.model.model_name
-        model_type = ModelType.Main
+        model_key = self.model.key
 
         # TODO: not found exceptions
-        if not context.services.model_manager.model_exists(
-            model_name=model_name,
-            base_model=base_model,
-            model_type=model_type,
-        ):
-            raise Exception(f"Unknown {base_model} {model_type} model: {model_name}")
+        if not context.services.model_manager.store.exists(model_key):
+            raise Exception(f"Unknown model: {model_key}")
 
         return SDXLRefinerModelLoaderOutput(
             unet=UNetField(
                 unet=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.UNet,
+                    key=model_key,
+                    submodel_type=SubModelType.UNet,
                 ),
                 scheduler=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.Scheduler,
+                    key=model_key,
+                    submodel_type=SubModelType.Scheduler,
                 ),
                 loras=[],
             ),
             clip2=ClipField(
                 tokenizer=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.Tokenizer2,
+                    key=model_key,
+                    submodel_type=SubModelType.Tokenizer2,
                 ),
                 text_encoder=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.TextEncoder2,
+                    key=model_key,
+                    submodel_type=SubModelType.TextEncoder2,
                 ),
                 loras=[],
                 skipped_layers=0,
             ),
             vae=VaeField(
                 vae=ModelInfo(
-                    model_name=model_name,
-                    base_model=base_model,
-                    model_type=model_type,
-                    submodel=SubModelType.Vae,
+                    key=model_key,
+                    submodel_type=SubModelType.Vae,
                 ),
             ),
         )

invokeai/app/invocations/t2i_adapter.py

@@ -1,6 +1,6 @@
 from typing import Union
 
-from pydantic import BaseModel, ConfigDict, Field
+from pydantic import BaseModel, Field, field_validator, model_validator
 
 from invokeai.app.invocations.baseinvocation import (
     BaseInvocation,
@@ -14,15 +14,12 @@ from invokeai.app.invocations.baseinvocation import (
 )
 from invokeai.app.invocations.controlnet_image_processors import CONTROLNET_RESIZE_VALUES
 from invokeai.app.invocations.primitives import ImageField
+from invokeai.app.invocations.util import validate_begin_end_step, validate_weights
 from invokeai.app.shared.fields import FieldDescriptions
-from invokeai.backend.model_management.models.base import BaseModelType
 
 
 class T2IAdapterModelField(BaseModel):
-    model_name: str = Field(description="Name of the T2I-Adapter model")
-    base_model: BaseModelType = Field(description="Base model")
-
-    model_config = ConfigDict(protected_namespaces=())
+    key: str = Field(description="Model record key for the T2I-Adapter model")
 
 
 class T2IAdapterField(BaseModel):
@@ -37,6 +34,17 @@ class T2IAdapterField(BaseModel):
     )
     resize_mode: CONTROLNET_RESIZE_VALUES = Field(default="just_resize", description="The resize mode to use")
 
+    @field_validator("weight")
+    @classmethod
+    def validate_ip_adapter_weight(cls, v):
+        validate_weights(v)
+        return v
+
+    @model_validator(mode="after")
+    def validate_begin_end_step_percent(self):
+        validate_begin_end_step(self.begin_step_percent, self.end_step_percent)
+        return self
+
 
 @invocation_output("t2i_adapter_output")
 class T2IAdapterOutput(BaseInvocationOutput):
@@ -44,7 +52,7 @@ class T2IAdapterOutput(BaseInvocationOutput):
 
 
 @invocation(
-    "t2i_adapter", title="T2I-Adapter", tags=["t2i_adapter", "control"], category="t2i_adapter", version="1.0.0"
+    "t2i_adapter", title="T2I-Adapter", tags=["t2i_adapter", "control"], category="t2i_adapter", version="1.0.1"
 )
 class T2IAdapterInvocation(BaseInvocation):
     """Collects T2I-Adapter info to pass to other nodes."""
@@ -61,7 +69,7 @@ class T2IAdapterInvocation(BaseInvocation):
         default=1, ge=0, description="The weight given to the T2I-Adapter", title="Weight"
     )
     begin_step_percent: float = InputField(
-        default=0, ge=-1, le=2, description="When the T2I-Adapter is first applied (% of total steps)"
+        default=0, ge=0, le=1, description="When the T2I-Adapter is first applied (% of total steps)"
     )
     end_step_percent: float = InputField(
         default=1, ge=0, le=1, description="When the T2I-Adapter is last applied (% of total steps)"
@@ -71,6 +79,17 @@ class T2IAdapterInvocation(BaseInvocation):
         description="The resize mode applied to the T2I-Adapter input image so that it matches the target output size.",
     )
 
+    @field_validator("weight")
+    @classmethod
+    def validate_ip_adapter_weight(cls, v):
+        validate_weights(v)
+        return v
+
+    @model_validator(mode="after")
+    def validate_begin_end_step_percent(self):
+        validate_begin_end_step(self.begin_step_percent, self.end_step_percent)
+        return self
+
     def invoke(self, context: InvocationContext) -> T2IAdapterOutput:
         return T2IAdapterOutput(
             t2i_adapter=T2IAdapterField(

invokeai/app/invocations/tiles.py

@@ -77,7 +77,7 @@ class CalculateImageTilesInvocation(BaseInvocation):
     title="Calculate Image Tiles Even Split",
     tags=["tiles"],
     category="tiles",
-    version="1.0.0",
+    version="1.1.0",
     classification=Classification.Beta,
 )
 class CalculateImageTilesEvenSplitInvocation(BaseInvocation):
@@ -97,11 +97,11 @@ class CalculateImageTilesEvenSplitInvocation(BaseInvocation):
         ge=1,
         description="Number of tiles to divide image into on the y axis",
     )
-    overlap_fraction: float = InputField(
-        default=0.25,
+    overlap: int = InputField(
+        default=128,
         ge=0,
-        lt=1,
-        description="Overlap between adjacent tiles as a fraction of the tile's dimensions (0-1)",
+        multiple_of=8,
+        description="The overlap, in pixels, between adjacent tiles.",
     )
 
     def invoke(self, context: InvocationContext) -> CalculateImageTilesOutput:
@@ -110,7 +110,7 @@ class CalculateImageTilesEvenSplitInvocation(BaseInvocation):
             image_width=self.image_width,
             num_tiles_x=self.num_tiles_x,
             num_tiles_y=self.num_tiles_y,
-            overlap_fraction=self.overlap_fraction,
+            overlap=self.overlap,
         )
         return CalculateImageTilesOutput(tiles=tiles)
 

invokeai/app/invocations/upscale.py

@@ -5,12 +5,12 @@ from typing import Literal
 import cv2
 import numpy as np
 import torch
-from basicsr.archs.rrdbnet_arch import RRDBNet
 from PIL import Image
 from pydantic import ConfigDict
 
 from invokeai.app.invocations.primitives import ImageField, ImageOutput
 from invokeai.app.services.image_records.image_records_common import ImageCategory, ResourceOrigin
+from invokeai.backend.image_util.basicsr.rrdbnet_arch import RRDBNet
 from invokeai.backend.image_util.realesrgan.realesrgan import RealESRGAN
 from invokeai.backend.util.devices import choose_torch_device
 

invokeai/app/invocations/util.py

@@ -0,0 +1,14 @@
+from typing import Union
+
+
+def validate_weights(weights: Union[float, list[float]]) -> None:
+    """Validate that all control weights in the valid range"""
+    to_validate = weights if isinstance(weights, list) else [weights]
+    if any(i < -1 or i > 2 for i in to_validate):
+        raise ValueError("Control weights must be within -1 to 2 range")
+
+
+def validate_begin_end_step(begin_step_percent: float, end_step_percent: float) -> None:
+    """Validate that begin_step_percent is less than end_step_percent"""
+    if begin_step_percent >= end_step_percent:
+        raise ValueError("Begin step percent must be less than or equal to end step percent")

invokeai/app/services/config/__init__.py

@@ -1,5 +1,7 @@
 """Init file for InvokeAI configure package."""
 
+from invokeai.app.services.config.config_common import PagingArgumentParser
+
 from .config_default import InvokeAIAppConfig, get_invokeai_config
 
-__all__ = ["InvokeAIAppConfig", "get_invokeai_config"]
+__all__ = ["InvokeAIAppConfig", "get_invokeai_config", "PagingArgumentParser"]

invokeai/app/services/config/config_base.py

@@ -27,11 +27,11 @@ class InvokeAISettings(BaseSettings):
     """Runtime configuration settings in which default values are read from an omegaconf .yaml file."""
 
     initconf: ClassVar[Optional[DictConfig]] = None
-    argparse_groups: ClassVar[Dict] = {}
+    argparse_groups: ClassVar[Dict[str, Any]] = {}
 
     model_config = SettingsConfigDict(env_file_encoding="utf-8", arbitrary_types_allowed=True, case_sensitive=True)
 
-    def parse_args(self, argv: Optional[list] = sys.argv[1:]):
+    def parse_args(self, argv: Optional[List[str]] = sys.argv[1:]) -> None:
         """Call to parse command-line arguments."""
         parser = self.get_parser()
         opt, unknown_opts = parser.parse_known_args(argv)
@@ -68,7 +68,7 @@ class InvokeAISettings(BaseSettings):
         return OmegaConf.to_yaml(conf)
 
     @classmethod
-    def add_parser_arguments(cls, parser):
+    def add_parser_arguments(cls, parser: ArgumentParser) -> None:
         """Dynamically create arguments for a settings parser."""
         if "type" in get_type_hints(cls):
             settings_stanza = get_args(get_type_hints(cls)["type"])[0]
@@ -117,7 +117,8 @@ class InvokeAISettings(BaseSettings):
         """Return the category of a setting."""
         hints = get_type_hints(cls)
         if command_field in hints:
-            return get_args(hints[command_field])[0]
+            result: str = get_args(hints[command_field])[0]
+            return result
         else:
             return "Uncategorized"
 
@@ -158,7 +159,7 @@ class InvokeAISettings(BaseSettings):
         ]
 
     @classmethod
-    def add_field_argument(cls, command_parser, name: str, field, default_override=None):
+    def add_field_argument(cls, command_parser, name: str, field, default_override=None) -> None:
         """Add the argparse arguments for a setting parser."""
         field_type = get_type_hints(cls).get(name)
         default = (

invokeai/app/services/config/config_common.py

@@ -21,7 +21,7 @@ class PagingArgumentParser(argparse.ArgumentParser):
     It also supports reading defaults from an init file.
     """
 
-    def print_help(self, file=None):
+    def print_help(self, file=None) -> None:
         text = self.format_help()
         pydoc.pager(text)
 

invokeai/app/services/config/config_default.py

@@ -173,10 +173,10 @@ from __future__ import annotations
 
 import os
 from pathlib import Path
-from typing import Any, ClassVar, Dict, List, Literal, Optional, Union, get_type_hints
+from typing import Any, ClassVar, Dict, List, Literal, Optional
 
 from omegaconf import DictConfig, OmegaConf
-from pydantic import Field, TypeAdapter
+from pydantic import Field
 from pydantic.config import JsonDict
 from pydantic_settings import SettingsConfigDict
 
@@ -185,7 +185,9 @@ from .config_base import InvokeAISettings
 INIT_FILE = Path("invokeai.yaml")
 DB_FILE = Path("invokeai.db")
 LEGACY_INIT_FILE = Path("invokeai.init")
-DEFAULT_MAX_VRAM = 0.5
+DEFAULT_RAM_CACHE = 10.0
+DEFAULT_VRAM_CACHE = 0.25
+DEFAULT_CONVERT_CACHE = 20.0
 
 
 class Categories(object):
@@ -209,7 +211,7 @@ class InvokeAIAppConfig(InvokeAISettings):
     """Configuration object for InvokeAI App."""
 
     singleton_config: ClassVar[Optional[InvokeAIAppConfig]] = None
-    singleton_init: ClassVar[Optional[Dict]] = None
+    singleton_init: ClassVar[Optional[Dict[str, Any]]] = None
 
     # fmt: off
     type: Literal["InvokeAI"] = "InvokeAI"
@@ -237,6 +239,7 @@ class InvokeAIAppConfig(InvokeAISettings):
     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)
     conf_path           : Path = Field(default=Path('configs/models.yaml'), description='Path to models definition file', 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)
@@ -251,26 +254,32 @@ class InvokeAIAppConfig(InvokeAISettings):
     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)
 
+    # 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)
+    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)
 
     version             : bool = Field(default=False, description="Show InvokeAI version and exit", json_schema_extra=Categories.Other)
 
     # CACHE
-    ram                 : float = Field(default=7.5, 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=0.25, 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 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, )
+    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, )
     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", "float32", "autocast"] = Field(default="auto", description="Floating point precision", json_schema_extra=Categories.Device)
+    precision           : Literal["auto", "float16", "bfloat16", "float32", "autocast"] = Field(default="auto", description="Floating point precision", 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=6, 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)
+    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)
 
     # QUEUE
     max_queue_size      : int = Field(default=10000, gt=0, description="Maximum number of items in the session queue", json_schema_extra=Categories.Queue)
@@ -280,6 +289,9 @@ class InvokeAIAppConfig(InvokeAISettings):
     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)
 
+    # 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)
+
     # 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)
@@ -289,6 +301,7 @@ class InvokeAIAppConfig(InvokeAISettings):
     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)
+
     # 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)
 
@@ -301,8 +314,8 @@ class InvokeAIAppConfig(InvokeAISettings):
         self,
         argv: Optional[list[str]] = None,
         conf: Optional[DictConfig] = None,
-        clobber=False,
-    ):
+        clobber: Optional[bool] = False,
+    ) -> None:
         """
         Update settings with contents of init file, environment, and command-line settings.
 
@@ -328,16 +341,12 @@ class InvokeAIAppConfig(InvokeAISettings):
         super().parse_args(argv)
 
         if self.singleton_init and not clobber:
-            hints = get_type_hints(self.__class__)
-            for k in self.singleton_init:
-                setattr(
-                    self,
-                    k,
-                    TypeAdapter(hints[k]).validate_python(self.singleton_init[k]),
-                )
+            # When setting values in this way, set validate_assignment to true if you want to validate the value.
+            for k, v in self.singleton_init.items():
+                setattr(self, k, v)
 
     @classmethod
-    def get_config(cls, **kwargs: Dict[str, Any]) -> InvokeAIAppConfig:
+    def get_config(cls, **kwargs: Any) -> InvokeAIAppConfig:
         """Return a singleton InvokeAIAppConfig configuration object."""
         if (
             cls.singleton_config is None
@@ -356,7 +365,7 @@ class InvokeAIAppConfig(InvokeAISettings):
         else:
             root = self.find_root().expanduser().absolute()
         self.root = root  # insulate ourselves from relative paths that may change
-        return root
+        return root.resolve()
 
     @property
     def root_dir(self) -> Path:
@@ -400,6 +409,11 @@ class InvokeAIAppConfig(InvokeAISettings):
         """Path to the models directory."""
         return self._resolve(self.models_dir)
 
+    @property
+    def models_convert_cache_path(self) -> Path:
+        """Path to the converted cache models directory."""
+        return self._resolve(self.convert_cache_dir)
+
     @property
     def custom_nodes_path(self) -> Path:
         """Path to the custom nodes directory."""
@@ -429,15 +443,20 @@ class InvokeAIAppConfig(InvokeAISettings):
         return True
 
     @property
-    def ram_cache_size(self) -> Union[Literal["auto"], float]:
-        """Return the ram cache size using the legacy or modern setting."""
+    def ram_cache_size(self) -> float:
+        """Return the ram cache size using the legacy or modern setting (GB)."""
         return self.max_cache_size or self.ram
 
     @property
-    def vram_cache_size(self) -> Union[Literal["auto"], float]:
-        """Return the vram cache size using the legacy or modern setting."""
+    def vram_cache_size(self) -> float:
+        """Return the vram cache size using the legacy or modern setting (GB)."""
         return self.max_vram_cache_size or self.vram
 
+    @property
+    def convert_cache_size(self) -> float:
+        """Return the convert cache size on disk (GB)."""
+        return self.convert_cache
+
     @property
     def use_cpu(self) -> bool:
         """Return true if the device is set to CPU or the always_use_cpu flag is set."""
@@ -449,13 +468,18 @@ class InvokeAIAppConfig(InvokeAISettings):
         disabled_in_config = not self.xformers_enabled
         return disabled_in_config and self.attention_type != "xformers"
 
+    @property
+    def profiles_path(self) -> Path:
+        """Path to the graph profiles directory."""
+        return self._resolve(self.profiles_dir)
+
     @staticmethod
     def find_root() -> Path:
         """Choose the runtime root directory when not specified on command line or init file."""
         return _find_root()
 
 
-def get_invokeai_config(**kwargs) -> InvokeAIAppConfig:
+def get_invokeai_config(**kwargs: Any) -> InvokeAIAppConfig:
     """Legacy function which returns InvokeAIAppConfig.get_config()."""
     return InvokeAIAppConfig.get_config(**kwargs)
 

invokeai/app/services/download/__init__.py

@@ -0,0 +1,12 @@
+"""Init file for download queue."""
+from .download_base import DownloadJob, DownloadJobStatus, DownloadQueueServiceBase, UnknownJobIDException
+from .download_default import DownloadQueueService, TqdmProgress
+
+__all__ = [
+    "DownloadJob",
+    "DownloadQueueServiceBase",
+    "DownloadQueueService",
+    "TqdmProgress",
+    "DownloadJobStatus",
+    "UnknownJobIDException",
+]

invokeai/app/services/download/download_base.py

@@ -0,0 +1,275 @@
+# Copyright (c) 2023 Lincoln D. Stein and the InvokeAI Development Team
+"""Model download service."""
+
+from abc import ABC, abstractmethod
+from enum import Enum
+from functools import total_ordering
+from pathlib import Path
+from typing import Any, Callable, List, Optional
+
+from pydantic import BaseModel, Field, PrivateAttr
+from pydantic.networks import AnyHttpUrl
+
+
+class DownloadJobStatus(str, Enum):
+    """State of a download job."""
+
+    WAITING = "waiting"  # not enqueued, will not run
+    RUNNING = "running"  # actively downloading
+    COMPLETED = "completed"  # finished running
+    CANCELLED = "cancelled"  # user cancelled
+    ERROR = "error"  # terminated with an error message
+
+
+class DownloadJobCancelledException(Exception):
+    """This exception is raised when a download job is cancelled."""
+
+
+class UnknownJobIDException(Exception):
+    """This exception is raised when an invalid job id is referened."""
+
+
+class ServiceInactiveException(Exception):
+    """This exception is raised when user attempts to initiate a download before the service is started."""
+
+
+DownloadEventHandler = Callable[["DownloadJob"], None]
+DownloadExceptionHandler = Callable[["DownloadJob", Optional[Exception]], None]
+
+
+@total_ordering
+class DownloadJob(BaseModel):
+    """Class to monitor and control a model download request."""
+
+    # required variables to be passed in on creation
+    source: AnyHttpUrl = Field(description="Where to download from. Specific types specified in child classes.")
+    dest: Path = Field(description="Destination of downloaded model on local disk; a directory or file path")
+    access_token: Optional[str] = Field(default=None, description="authorization token for protected resources")
+    # automatically assigned on creation
+    id: int = Field(description="Numeric ID of this job", default=-1)  # default id is a sentinel
+    priority: int = Field(default=10, description="Queue priority; lower values are higher priority")
+
+    # set internally during download process
+    status: DownloadJobStatus = Field(default=DownloadJobStatus.WAITING, description="Status of the download")
+    download_path: Optional[Path] = Field(default=None, description="Final location of downloaded file")
+    job_started: Optional[str] = Field(default=None, description="Timestamp for when the download job started")
+    job_ended: Optional[str] = Field(
+        default=None, description="Timestamp for when the download job ende1d (completed or errored)"
+    )
+    content_type: Optional[str] = Field(default=None, description="Content type of downloaded file")
+    bytes: int = Field(default=0, description="Bytes downloaded so far")
+    total_bytes: int = Field(default=0, description="Total file size (bytes)")
+
+    # set when an error occurs
+    error_type: Optional[str] = Field(default=None, description="Name of exception that caused an error")
+    error: Optional[str] = Field(default=None, description="Traceback of the exception that caused an error")
+
+    # internal flag
+    _cancelled: bool = PrivateAttr(default=False)
+
+    # optional event handlers passed in on creation
+    _on_start: Optional[DownloadEventHandler] = PrivateAttr(default=None)
+    _on_progress: Optional[DownloadEventHandler] = PrivateAttr(default=None)
+    _on_complete: Optional[DownloadEventHandler] = PrivateAttr(default=None)
+    _on_cancelled: Optional[DownloadEventHandler] = PrivateAttr(default=None)
+    _on_error: Optional[DownloadExceptionHandler] = PrivateAttr(default=None)
+
+    def __hash__(self) -> int:
+        """Return hash of the string representation of this object, for indexing."""
+        return hash(str(self))
+
+    def __le__(self, other: "DownloadJob") -> bool:
+        """Return True if this job's priority is less than another's."""
+        return self.priority <= other.priority
+
+    def cancel(self) -> None:
+        """Call to cancel the job."""
+        self._cancelled = True
+
+    # cancelled and the callbacks are private attributes in order to prevent
+    # them from being serialized and/or used in the Json Schema
+    @property
+    def cancelled(self) -> bool:
+        """Call to cancel the job."""
+        return self._cancelled
+
+    @property
+    def complete(self) -> bool:
+        """Return true if job completed without errors."""
+        return self.status == DownloadJobStatus.COMPLETED
+
+    @property
+    def running(self) -> bool:
+        """Return true if the job is running."""
+        return self.status == DownloadJobStatus.RUNNING
+
+    @property
+    def errored(self) -> bool:
+        """Return true if the job is errored."""
+        return self.status == DownloadJobStatus.ERROR
+
+    @property
+    def in_terminal_state(self) -> bool:
+        """Return true if job has finished, one way or another."""
+        return self.status not in [DownloadJobStatus.WAITING, DownloadJobStatus.RUNNING]
+
+    @property
+    def on_start(self) -> Optional[DownloadEventHandler]:
+        """Return the on_start event handler."""
+        return self._on_start
+
+    @property
+    def on_progress(self) -> Optional[DownloadEventHandler]:
+        """Return the on_progress event handler."""
+        return self._on_progress
+
+    @property
+    def on_complete(self) -> Optional[DownloadEventHandler]:
+        """Return the on_complete event handler."""
+        return self._on_complete
+
+    @property
+    def on_error(self) -> Optional[DownloadExceptionHandler]:
+        """Return the on_error event handler."""
+        return self._on_error
+
+    @property
+    def on_cancelled(self) -> Optional[DownloadEventHandler]:
+        """Return the on_cancelled event handler."""
+        return self._on_cancelled
+
+    def set_callbacks(
+        self,
+        on_start: Optional[DownloadEventHandler] = None,
+        on_progress: Optional[DownloadEventHandler] = None,
+        on_complete: Optional[DownloadEventHandler] = None,
+        on_cancelled: Optional[DownloadEventHandler] = None,
+        on_error: Optional[DownloadExceptionHandler] = None,
+    ) -> None:
+        """Set the callbacks for download events."""
+        self._on_start = on_start
+        self._on_progress = on_progress
+        self._on_complete = on_complete
+        self._on_error = on_error
+        self._on_cancelled = on_cancelled
+
+
+class DownloadQueueServiceBase(ABC):
+    """Multithreaded queue for downloading models via URL."""
+
+    @abstractmethod
+    def start(self, *args: Any, **kwargs: Any) -> None:
+        """Start the download worker threads."""
+
+    @abstractmethod
+    def stop(self, *args: Any, **kwargs: Any) -> None:
+        """Stop the download worker threads."""
+
+    @abstractmethod
+    def download(
+        self,
+        source: AnyHttpUrl,
+        dest: Path,
+        priority: int = 10,
+        access_token: Optional[str] = None,
+        on_start: Optional[DownloadEventHandler] = None,
+        on_progress: Optional[DownloadEventHandler] = None,
+        on_complete: Optional[DownloadEventHandler] = None,
+        on_cancelled: Optional[DownloadEventHandler] = None,
+        on_error: Optional[DownloadExceptionHandler] = None,
+    ) -> DownloadJob:
+        """
+        Create and enqueue download job.
+
+        :param source: Source of the download as a URL.
+        :param dest: Path to download to. See below.
+        :param on_start, on_progress, on_complete, on_error: Callbacks for the indicated
+         events.
+        :returns: A DownloadJob object for monitoring the state of the download.
+
+        The `dest` argument is a Path object. Its behavior is:
+
+        1. If the path exists and is a directory, then the URL contents will be downloaded
+            into that directory using the filename indicated in the response's `Content-Disposition` field.
+            If no content-disposition is present, then the last component of the URL will be used (similar to
+            wget's behavior).
+        2. If the path does not exist, then it is taken as the name of a new file to create with the downloaded
+           content.
+        3. If the path exists and is an existing file, then the downloader will try to resume the download from
+           the end of the existing file.
+
+        """
+        pass
+
+    @abstractmethod
+    def submit_download_job(
+        self,
+        job: DownloadJob,
+        on_start: Optional[DownloadEventHandler] = None,
+        on_progress: Optional[DownloadEventHandler] = None,
+        on_complete: Optional[DownloadEventHandler] = None,
+        on_cancelled: Optional[DownloadEventHandler] = None,
+        on_error: Optional[DownloadExceptionHandler] = None,
+    ) -> None:
+        """
+        Enqueue a download job.
+
+        :param job: The DownloadJob
+        :param on_start, on_progress, on_complete, on_error: Callbacks for the indicated
+         events.
+        """
+        pass
+
+    @abstractmethod
+    def list_jobs(self) -> List[DownloadJob]:
+        """
+        List active download jobs.
+
+        :returns List[DownloadJob]: List of download jobs whose state is not "completed."
+        """
+        pass
+
+    @abstractmethod
+    def id_to_job(self, id: int) -> DownloadJob:
+        """
+        Return the DownloadJob corresponding to the integer ID.
+
+        :param id: ID of the DownloadJob.
+
+        Exceptions:
+        * UnknownJobIDException
+        """
+        pass
+
+    @abstractmethod
+    def cancel_all_jobs(self) -> None:
+        """Cancel all active and enquedjobs."""
+        pass
+
+    @abstractmethod
+    def prune_jobs(self) -> None:
+        """Prune completed and errored queue items from the job list."""
+        pass
+
+    @abstractmethod
+    def cancel_job(self, job: DownloadJob) -> None:
+        """Cancel the job, clearing partial downloads and putting it into ERROR state."""
+        pass
+
+    @abstractmethod
+    def join(self) -> None:
+        """Wait until all jobs are off the queue."""
+        pass
+
+    @abstractmethod
+    def wait_for_job(self, job: DownloadJob, timeout: int = 0) -> DownloadJob:
+        """Wait until the indicated download job has reached a terminal state.
+
+        This will block until the indicated install job has completed,
+        been cancelled, or errored out.
+
+        :param job: The job to wait on.
+        :param timeout: Wait up to indicated number of seconds. Raise a TimeoutError if
+        the job hasn't completed within the indicated time.
+        """
+        pass

invokeai/app/services/download/download_default.py

@@ -0,0 +1,449 @@
+# Copyright (c) 2023, Lincoln D. Stein
+"""Implementation of multithreaded download queue for invokeai."""
+
+import os
+import re
+import threading
+import time
+import traceback
+from pathlib import Path
+from queue import Empty, PriorityQueue
+from typing import Any, Dict, List, Optional, Set
+
+import requests
+from pydantic.networks import AnyHttpUrl
+from requests import HTTPError
+from tqdm import tqdm
+
+from invokeai.app.services.events.events_base import EventServiceBase
+from invokeai.app.util.misc import get_iso_timestamp
+from invokeai.backend.util.logging import InvokeAILogger
+
+from .download_base import (
+    DownloadEventHandler,
+    DownloadExceptionHandler,
+    DownloadJob,
+    DownloadJobCancelledException,
+    DownloadJobStatus,
+    DownloadQueueServiceBase,
+    ServiceInactiveException,
+    UnknownJobIDException,
+)
+
+# Maximum number of bytes to download during each call to requests.iter_content()
+DOWNLOAD_CHUNK_SIZE = 100000
+
+
+class DownloadQueueService(DownloadQueueServiceBase):
+    """Class for queued download of models."""
+
+    def __init__(
+        self,
+        max_parallel_dl: int = 5,
+        event_bus: Optional[EventServiceBase] = None,
+        requests_session: Optional[requests.sessions.Session] = None,
+    ):
+        """
+        Initialize DownloadQueue.
+
+        :param max_parallel_dl: Number of simultaneous downloads allowed [5].
+        :param requests_session: Optional requests.sessions.Session object, for unit tests.
+        """
+        self._jobs: Dict[int, DownloadJob] = {}
+        self._next_job_id = 0
+        self._queue: PriorityQueue[DownloadJob] = PriorityQueue()
+        self._stop_event = threading.Event()
+        self._job_completed_event = threading.Event()
+        self._worker_pool: Set[threading.Thread] = set()
+        self._lock = threading.Lock()
+        self._logger = InvokeAILogger.get_logger("DownloadQueueService")
+        self._event_bus = event_bus
+        self._requests = requests_session or requests.Session()
+        self._accept_download_requests = False
+        self._max_parallel_dl = max_parallel_dl
+
+    def start(self, *args: Any, **kwargs: Any) -> None:
+        """Start the download worker threads."""
+        with self._lock:
+            if self._worker_pool:
+                raise Exception("Attempt to start the download service twice")
+            self._stop_event.clear()
+            self._start_workers(self._max_parallel_dl)
+            self._accept_download_requests = True
+
+    def stop(self, *args: Any, **kwargs: Any) -> None:
+        """Stop the download worker threads."""
+        with self._lock:
+            if not self._worker_pool:
+                raise Exception("Attempt to stop the download service before it was started")
+            self._accept_download_requests = False  # reject attempts to add new jobs to queue
+            queued_jobs = [x for x in self.list_jobs() if x.status == DownloadJobStatus.WAITING]
+            active_jobs = [x for x in self.list_jobs() if x.status == DownloadJobStatus.RUNNING]
+            if queued_jobs:
+                self._logger.warning(f"Cancelling {len(queued_jobs)} queued downloads")
+            if active_jobs:
+                self._logger.info(f"Waiting for {len(active_jobs)} active download jobs to complete")
+            with self._queue.mutex:
+                self._queue.queue.clear()
+            self.join()  # wait for all active jobs to finish
+            self._stop_event.set()
+            self._worker_pool.clear()
+
+    def submit_download_job(
+        self,
+        job: DownloadJob,
+        on_start: Optional[DownloadEventHandler] = None,
+        on_progress: Optional[DownloadEventHandler] = None,
+        on_complete: Optional[DownloadEventHandler] = None,
+        on_cancelled: Optional[DownloadEventHandler] = None,
+        on_error: Optional[DownloadExceptionHandler] = None,
+    ) -> None:
+        """Enqueue a download job."""
+        if not self._accept_download_requests:
+            raise ServiceInactiveException(
+                "The download service is not currently accepting requests. Please call start() to initialize the service."
+            )
+        with self._lock:
+            job.id = self._next_job_id
+            self._next_job_id += 1
+            job.set_callbacks(
+                on_start=on_start,
+                on_progress=on_progress,
+                on_complete=on_complete,
+                on_cancelled=on_cancelled,
+                on_error=on_error,
+            )
+            self._jobs[job.id] = job
+            self._queue.put(job)
+
+    def download(
+        self,
+        source: AnyHttpUrl,
+        dest: Path,
+        priority: int = 10,
+        access_token: Optional[str] = None,
+        on_start: Optional[DownloadEventHandler] = None,
+        on_progress: Optional[DownloadEventHandler] = None,
+        on_complete: Optional[DownloadEventHandler] = None,
+        on_cancelled: Optional[DownloadEventHandler] = None,
+        on_error: Optional[DownloadExceptionHandler] = None,
+    ) -> DownloadJob:
+        """Create and enqueue a download job and return it."""
+        if not self._accept_download_requests:
+            raise ServiceInactiveException(
+                "The download service is not currently accepting requests. Please call start() to initialize the service."
+            )
+        job = DownloadJob(
+            source=source,
+            dest=dest,
+            priority=priority,
+            access_token=access_token,
+        )
+        self.submit_download_job(
+            job,
+            on_start=on_start,
+            on_progress=on_progress,
+            on_complete=on_complete,
+            on_cancelled=on_cancelled,
+            on_error=on_error,
+        )
+        return job
+
+    def join(self) -> None:
+        """Wait for all jobs to complete."""
+        self._queue.join()
+
+    def list_jobs(self) -> List[DownloadJob]:
+        """List all the jobs."""
+        return list(self._jobs.values())
+
+    def prune_jobs(self) -> None:
+        """Prune completed and errored queue items from the job list."""
+        with self._lock:
+            to_delete = set()
+            for job_id, job in self._jobs.items():
+                if job.in_terminal_state:
+                    to_delete.add(job_id)
+            for job_id in to_delete:
+                del self._jobs[job_id]
+
+    def id_to_job(self, id: int) -> DownloadJob:
+        """Translate a job ID into a DownloadJob object."""
+        try:
+            return self._jobs[id]
+        except KeyError as excp:
+            raise UnknownJobIDException("Unrecognized job") from excp
+
+    def cancel_job(self, job: DownloadJob) -> None:
+        """
+        Cancel the indicated job.
+
+        If it is running it will be stopped.
+        job.status will be set to DownloadJobStatus.CANCELLED
+        """
+        with self._lock:
+            job.cancel()
+
+    def cancel_all_jobs(self) -> None:
+        """Cancel all jobs (those not in enqueued, running or paused state)."""
+        for job in self._jobs.values():
+            if not job.in_terminal_state:
+                self.cancel_job(job)
+
+    def wait_for_job(self, job: DownloadJob, timeout: int = 0) -> DownloadJob:
+        """Block until the indicated job has reached terminal state, or when timeout limit reached."""
+        start = time.time()
+        while not job.in_terminal_state:
+            if self._job_completed_event.wait(timeout=0.25):  # in case we miss an event
+                self._job_completed_event.clear()
+            if timeout > 0 and time.time() - start > timeout:
+                raise TimeoutError("Timeout exceeded")
+        return job
+
+    def _start_workers(self, max_workers: int) -> None:
+        """Start the requested number of worker threads."""
+        self._stop_event.clear()
+        for i in range(0, max_workers):  # noqa B007
+            worker = threading.Thread(target=self._download_next_item, daemon=True)
+            self._logger.debug(f"Download queue worker thread {worker.name} starting.")
+            worker.start()
+            self._worker_pool.add(worker)
+
+    def _download_next_item(self) -> None:
+        """Worker thread gets next job on priority queue."""
+        done = False
+        while not done:
+            if self._stop_event.is_set():
+                done = True
+                continue
+            try:
+                job = self._queue.get(timeout=1)
+            except Empty:
+                continue
+            try:
+                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()
+                self._signal_job_error(job, excp)
+            except DownloadJobCancelledException:
+                self._signal_job_cancelled(job)
+                self._cleanup_cancelled_job(job)
+
+            finally:
+                job.job_ended = get_iso_timestamp()
+                self._job_completed_event.set()  # signal a change to terminal state
+                self._queue.task_done()
+        self._logger.debug(f"Download queue worker thread {threading.current_thread().name} exiting.")
+
+    def _do_download(self, job: DownloadJob) -> None:
+        """Do the actual download."""
+        url = job.source
+        header = {"Authorization": f"Bearer {job.access_token}"} if job.access_token else {}
+        open_mode = "wb"
+
+        # Make a streaming request. This will retrieve headers including
+        # content-length and content-disposition, but not fetch any content itself
+        resp = self._requests.get(str(url), headers=header, stream=True)
+        if not resp.ok:
+            raise HTTPError(resp.reason)
+
+        job.content_type = resp.headers.get("Content-Type")
+        content_length = int(resp.headers.get("content-length", 0))
+        job.total_bytes = content_length
+
+        if job.dest.is_dir():
+            file_name = os.path.basename(str(url.path))  # default is to use the last bit of the URL
+
+            if match := re.search('filename="(.+)"', resp.headers.get("Content-Disposition", "")):
+                remote_name = match.group(1)
+                if self._validate_filename(job.dest.as_posix(), remote_name):
+                    file_name = remote_name
+
+            job.download_path = job.dest / file_name
+
+        else:
+            job.dest.parent.mkdir(parents=True, exist_ok=True)
+            job.download_path = job.dest
+
+        assert job.download_path
+
+        # Don't clobber an existing file. See commit 82c2c85202f88c6d24ff84710f297cfc6ae174af
+        # for code that instead resumes an interrupted download.
+        if job.download_path.exists():
+            raise OSError(f"[Errno 17] File {job.download_path} exists")
+
+        # append ".downloading" to the path
+        in_progress_path = self._in_progress_path(job.download_path)
+
+        # signal caller that the download is starting. At this point, key fields such as
+        # download_path and total_bytes will be populated. We call it here because the might
+        # discover that the local file is already complete and generate a COMPLETED status.
+        self._signal_job_started(job)
+
+        # "range not satisfiable" - local file is at least as large as the remote file
+        if resp.status_code == 416 or (content_length > 0 and job.bytes >= content_length):
+            self._logger.warning(f"{job.download_path}: complete file found. Skipping.")
+            return
+
+        # "partial content" - local file is smaller than remote file
+        elif resp.status_code == 206 or job.bytes > 0:
+            self._logger.warning(f"{job.download_path}: partial file found. Resuming")
+
+        # some other error
+        elif resp.status_code != 200:
+            raise HTTPError(resp.reason)
+
+        self._logger.debug(f"{job.source}: Downloading {job.download_path}")
+        report_delta = job.total_bytes / 100  # report every 1% change
+        last_report_bytes = 0
+
+        # DOWNLOAD LOOP
+        with open(in_progress_path, open_mode) as file:
+            for data in resp.iter_content(chunk_size=DOWNLOAD_CHUNK_SIZE):
+                if job.cancelled:
+                    raise DownloadJobCancelledException("Job was cancelled at caller's request")
+
+                job.bytes += file.write(data)
+                if (job.bytes - last_report_bytes >= report_delta) or (job.bytes >= job.total_bytes):
+                    last_report_bytes = job.bytes
+                    self._signal_job_progress(job)
+
+        # if we get here we are done and can rename the file to the original dest
+        self._logger.debug(f"{job.source}: saved to {job.download_path} (bytes={job.bytes})")
+        in_progress_path.rename(job.download_path)
+
+    def _validate_filename(self, directory: str, filename: str) -> bool:
+        pc_name_max = os.pathconf(directory, "PC_NAME_MAX") if hasattr(os, "pathconf") else 260  # hardcoded for windows
+        pc_path_max = (
+            os.pathconf(directory, "PC_PATH_MAX") if hasattr(os, "pathconf") else 32767
+        )  # hardcoded for windows with long names enabled
+        if "/" in filename:
+            return False
+        if filename.startswith(".."):
+            return False
+        if len(filename) > pc_name_max:
+            return False
+        if len(os.path.join(directory, filename)) > pc_path_max:
+            return False
+        return True
+
+    def _in_progress_path(self, path: Path) -> Path:
+        return path.with_name(path.name + ".downloading")
+
+    def _signal_job_started(self, job: DownloadJob) -> None:
+        job.status = DownloadJobStatus.RUNNING
+        if job.on_start:
+            try:
+                job.on_start(job)
+            except Exception as e:
+                self._logger.error(
+                    f"An error occurred while processing the on_start callback: {traceback.format_exception(e)}"
+                )
+        if self._event_bus:
+            assert job.download_path
+            self._event_bus.emit_download_started(str(job.source), job.download_path.as_posix())
+
+    def _signal_job_progress(self, job: DownloadJob) -> None:
+        if job.on_progress:
+            try:
+                job.on_progress(job)
+            except Exception as e:
+                self._logger.error(
+                    f"An error occurred while processing the on_progress callback: {traceback.format_exception(e)}"
+                )
+        if self._event_bus:
+            assert job.download_path
+            self._event_bus.emit_download_progress(
+                str(job.source),
+                download_path=job.download_path.as_posix(),
+                current_bytes=job.bytes,
+                total_bytes=job.total_bytes,
+            )
+
+    def _signal_job_complete(self, job: DownloadJob) -> None:
+        job.status = DownloadJobStatus.COMPLETED
+        if job.on_complete:
+            try:
+                job.on_complete(job)
+            except Exception as e:
+                self._logger.error(
+                    f"An error occurred while processing the on_complete callback: {traceback.format_exception(e)}"
+                )
+        if self._event_bus:
+            assert job.download_path
+            self._event_bus.emit_download_complete(
+                str(job.source), download_path=job.download_path.as_posix(), total_bytes=job.total_bytes
+            )
+
+    def _signal_job_cancelled(self, job: DownloadJob) -> None:
+        if job.status not in [DownloadJobStatus.RUNNING, DownloadJobStatus.WAITING]:
+            return
+        job.status = DownloadJobStatus.CANCELLED
+        if job.on_cancelled:
+            try:
+                job.on_cancelled(job)
+            except Exception as e:
+                self._logger.error(
+                    f"An error occurred while processing the on_cancelled callback: {traceback.format_exception(e)}"
+                )
+        if self._event_bus:
+            self._event_bus.emit_download_cancelled(str(job.source))
+
+    def _signal_job_error(self, job: DownloadJob, excp: Optional[Exception] = None) -> None:
+        job.status = DownloadJobStatus.ERROR
+        self._logger.error(f"{str(job.source)}: {traceback.format_exception(excp)}")
+        if job.on_error:
+            try:
+                job.on_error(job, excp)
+            except Exception as e:
+                self._logger.error(
+                    f"An error occurred while processing the on_error callback: {traceback.format_exception(e)}"
+                )
+        if self._event_bus:
+            assert job.error_type
+            assert job.error
+            self._event_bus.emit_download_error(str(job.source), error_type=job.error_type, error=job.error)
+
+    def _cleanup_cancelled_job(self, job: DownloadJob) -> None:
+        self._logger.debug(f"Cleaning up leftover files from cancelled download job {job.download_path}")
+        try:
+            if job.download_path:
+                partial_file = self._in_progress_path(job.download_path)
+                partial_file.unlink()
+        except OSError as excp:
+            self._logger.warning(excp)
+
+
+# Example on_progress event handler to display a TQDM status bar
+# Activate with:
+#   download_service.download(DownloadJob('http://foo.bar/baz', '/tmp', on_progress=TqdmProgress().update))
+class TqdmProgress(object):
+    """TQDM-based progress bar object to use in on_progress handlers."""
+
+    _bars: Dict[int, tqdm]  # type: ignore
+    _last: Dict[int, int]  # last bytes downloaded
+
+    def __init__(self) -> None:  # noqa D107
+        self._bars = {}
+        self._last = {}
+
+    def update(self, job: DownloadJob) -> None:  # noqa D102
+        job_id = job.id
+        # new job
+        if job_id not in self._bars:
+            assert job.download_path
+            dest = Path(job.download_path).name
+            self._bars[job_id] = tqdm(
+                desc=dest,
+                initial=0,
+                total=job.total_bytes,
+                unit="iB",
+                unit_scale=True,
+            )
+            self._last[job_id] = 0
+        self._bars[job_id].update(job.bytes - self._last[job_id])
+        self._last[job_id] = job.bytes

invokeai/app/services/events/events_base.py

@@ -1,7 +1,7 @@
 # Copyright (c) 2022 Kyle Schouviller (https://github.com/kyle0654)
 
 
-from typing import Any, Optional
+from typing import Any, Dict, List, Optional, Union
 
 from invokeai.app.services.invocation_processor.invocation_processor_common import ProgressImage
 from invokeai.app.services.session_queue.session_queue_common import (
@@ -11,12 +11,12 @@ from invokeai.app.services.session_queue.session_queue_common import (
     SessionQueueStatus,
 )
 from invokeai.app.util.misc import get_timestamp
-from invokeai.backend.model_management.model_manager import ModelInfo
-from invokeai.backend.model_management.models.base import BaseModelType, ModelType, SubModelType
+from invokeai.backend.model_manager import AnyModelConfig
 
 
 class EventServiceBase:
     queue_event: str = "queue_event"
+    download_event: str = "download_event"
     model_event: str = "model_event"
 
     """Basic event bus, to have an empty stand-in when not needed"""
@@ -32,6 +32,13 @@ class EventServiceBase:
             payload={"event": event_name, "data": payload},
         )
 
+    def __emit_download_event(self, event_name: str, payload: dict) -> None:
+        payload["timestamp"] = get_timestamp()
+        self.dispatch(
+            event_name=EventServiceBase.download_event,
+            payload={"event": event_name, "data": payload},
+        )
+
     def __emit_model_event(self, event_name: str, payload: dict) -> None:
         payload["timestamp"] = get_timestamp()
         self.dispatch(
@@ -163,10 +170,7 @@ class EventServiceBase:
         queue_item_id: int,
         queue_batch_id: str,
         graph_execution_state_id: str,
-        model_name: str,
-        base_model: BaseModelType,
-        model_type: ModelType,
-        submodel: SubModelType,
+        model_config: AnyModelConfig,
     ) -> None:
         """Emitted when a model is requested"""
         self.__emit_queue_event(
@@ -176,10 +180,7 @@ class EventServiceBase:
                 "queue_item_id": queue_item_id,
                 "queue_batch_id": queue_batch_id,
                 "graph_execution_state_id": graph_execution_state_id,
-                "model_name": model_name,
-                "base_model": base_model,
-                "model_type": model_type,
-                "submodel": submodel,
+                "model_config": model_config.model_dump(),
             },
         )
 
@@ -189,11 +190,7 @@ class EventServiceBase:
         queue_item_id: int,
         queue_batch_id: str,
         graph_execution_state_id: str,
-        model_name: str,
-        base_model: BaseModelType,
-        model_type: ModelType,
-        submodel: SubModelType,
-        model_info: ModelInfo,
+        model_config: AnyModelConfig,
     ) -> None:
         """Emitted when a model is correctly loaded (returns model info)"""
         self.__emit_queue_event(
@@ -203,13 +200,7 @@ class EventServiceBase:
                 "queue_item_id": queue_item_id,
                 "queue_batch_id": queue_batch_id,
                 "graph_execution_state_id": graph_execution_state_id,
-                "model_name": model_name,
-                "base_model": base_model,
-                "model_type": model_type,
-                "submodel": submodel,
-                "hash": model_info.hash,
-                "location": str(model_info.location),
-                "precision": str(model_info.precision),
+                "model_config": model_config.model_dump(),
             },
         )
 
@@ -323,53 +314,145 @@ class EventServiceBase:
             payload={"queue_id": queue_id},
         )
 
-    def emit_model_install_started(self, source: str) -> None:
+    def emit_download_started(self, source: str, download_path: str) -> None:
         """
-        Emitted when an install job is started.
+        Emit when a download job is started.
+
+        :param url: The downloaded url
+        """
+        self.__emit_download_event(
+            event_name="download_started",
+            payload={"source": source, "download_path": download_path},
+        )
+
+    def emit_download_progress(self, source: str, download_path: str, current_bytes: int, total_bytes: int) -> None:
+        """
+        Emit "download_progress" events at regular intervals during a download job.
+
+        :param source: The downloaded source
+        :param download_path: The local downloaded file
+        :param current_bytes: Number of bytes downloaded so far
+        :param total_bytes: The size of the file being downloaded (if known)
+        """
+        self.__emit_download_event(
+            event_name="download_progress",
+            payload={
+                "source": source,
+                "download_path": download_path,
+                "current_bytes": current_bytes,
+                "total_bytes": total_bytes,
+            },
+        )
+
+    def emit_download_complete(self, source: str, download_path: str, total_bytes: int) -> None:
+        """
+        Emit a "download_complete" event at the end of a successful download.
+
+        :param source: Source URL
+        :param download_path: Path to the locally downloaded file
+        :param total_bytes: The size of the downloaded file
+        """
+        self.__emit_download_event(
+            event_name="download_complete",
+            payload={
+                "source": source,
+                "download_path": download_path,
+                "total_bytes": total_bytes,
+            },
+        )
+
+    def emit_download_cancelled(self, source: str) -> None:
+        """Emit a "download_cancelled" event in the event that the download was cancelled by user."""
+        self.__emit_download_event(
+            event_name="download_cancelled",
+            payload={
+                "source": source,
+            },
+        )
+
+    def emit_download_error(self, source: str, error_type: str, error: str) -> None:
+        """
+        Emit a "download_error" event when an download job encounters an exception.
+
+        :param source: Source URL
+        :param error_type: The name of the exception that raised the error
+        :param error: The traceback from this error
+        """
+        self.__emit_download_event(
+            event_name="download_error",
+            payload={
+                "source": source,
+                "error_type": error_type,
+                "error": error,
+            },
+        )
+
+    def emit_model_install_downloading(
+        self,
+        source: str,
+        local_path: str,
+        bytes: int,
+        total_bytes: int,
+        parts: List[Dict[str, Union[str, int]]],
+    ) -> None:
+        """
+        Emit at intervals while the install job is in progress (remote models only).
+
+        :param source: Source of the model
+        :param local_path: Where model is downloading to
+        :param parts: Progress of downloading URLs that comprise the model, if any.
+        :param bytes: Number of bytes downloaded so far.
+        :param total_bytes: Total size of download, including all files.
+        This emits a Dict with keys "source", "local_path", "bytes" and "total_bytes".
+        """
+        self.__emit_model_event(
+            event_name="model_install_downloading",
+            payload={
+                "source": source,
+                "local_path": local_path,
+                "bytes": bytes,
+                "total_bytes": total_bytes,
+                "parts": parts,
+            },
+        )
+
+    def emit_model_install_running(self, source: str) -> None:
+        """
+        Emit once when an install job becomes active.
 
         :param source: Source of the model; local path, repo_id or url
         """
         self.__emit_model_event(
-            event_name="model_install_started",
+            event_name="model_install_running",
             payload={"source": source},
         )
 
-    def emit_model_install_completed(self, source: str, key: str) -> None:
+    def emit_model_install_completed(self, source: str, key: str, total_bytes: Optional[int] = None) -> None:
         """
-        Emitted when an install job is completed successfully.
+        Emit when an install job is completed successfully.
 
         :param source: Source of the model; local path, repo_id or url
         :param key: Model config record key
+        :param total_bytes: Size of the model (may be None for installation of a local path)
         """
         self.__emit_model_event(
             event_name="model_install_completed",
             payload={
                 "source": source,
+                "total_bytes": total_bytes,
                 "key": key,
             },
         )
 
-    def emit_model_install_progress(
-        self,
-        source: str,
-        current_bytes: int,
-        total_bytes: int,
-    ) -> None:
+    def emit_model_install_cancelled(self, source: str) -> None:
         """
-        Emitted while the install job is in progress.
-        (Downloaded models only)
+        Emit when an install job is cancelled.
 
-        :param source: Source of the model
-        :param current_bytes: Number of bytes downloaded so far
-        :param total_bytes: Total bytes to download
+        :param source: Source of the model; local path, repo_id or url
         """
         self.__emit_model_event(
-            event_name="model_install_progress",
-            payload={
-                "source": source,
-                "current_bytes": int,
-                "total_bytes": int,
-            },
+            event_name="model_install_cancelled",
+            payload={"source": source},
         )
 
     def emit_model_install_error(
@@ -379,10 +462,11 @@ class EventServiceBase:
         error: str,
     ) -> None:
         """
-        Emitted when an install job encounters an exception.
+        Emit when an install job encounters an exception.
 
         :param source: Source of the model
-        :param exception: The exception that raised the error
+        :param error_type: The name of the exception
+        :param error: A text description of the exception
         """
         self.__emit_model_event(
             event_name="model_install_error",

invokeai/app/services/invocation_processor/invocation_processor_default.py

@@ -1,11 +1,16 @@
 import time
 import traceback
+from contextlib import suppress
 from threading import BoundedSemaphore, Event, Thread
 from typing import Optional
 
 import invokeai.backend.util.logging as logger
 from invokeai.app.invocations.baseinvocation import InvocationContext
 from invokeai.app.services.invocation_queue.invocation_queue_common import InvocationQueueItem
+from invokeai.app.services.invocation_stats.invocation_stats_common import (
+    GESStatsNotFoundError,
+)
+from invokeai.app.util.profiler import Profiler
 
 from ..invoker import Invoker
 from .invocation_processor_base import InvocationProcessorABC
@@ -18,7 +23,7 @@ class DefaultInvocationProcessor(InvocationProcessorABC):
     __invoker: Invoker
     __threadLimit: BoundedSemaphore
 
-    def start(self, invoker) -> None:
+    def start(self, invoker: Invoker) -> None:
         # if we do want multithreading at some point, we could make this configurable
         self.__threadLimit = BoundedSemaphore(1)
         self.__invoker = invoker
@@ -39,6 +44,27 @@ class DefaultInvocationProcessor(InvocationProcessorABC):
             self.__threadLimit.acquire()
             queue_item: Optional[InvocationQueueItem] = None
 
+            profiler = (
+                Profiler(
+                    logger=self.__invoker.services.logger,
+                    output_dir=self.__invoker.services.configuration.profiles_path,
+                    prefix=self.__invoker.services.configuration.profile_prefix,
+                )
+                if self.__invoker.services.configuration.profile_graphs
+                else None
+            )
+
+            def stats_cleanup(graph_execution_state_id: str) -> None:
+                if profiler:
+                    profile_path = profiler.stop()
+                    stats_path = profile_path.with_suffix(".json")
+                    self.__invoker.services.performance_statistics.dump_stats(
+                        graph_execution_state_id=graph_execution_state_id, output_path=stats_path
+                    )
+                with suppress(GESStatsNotFoundError):
+                    self.__invoker.services.performance_statistics.log_stats(graph_execution_state_id)
+                    self.__invoker.services.performance_statistics.reset_stats(graph_execution_state_id)
+
             while not stop_event.is_set():
                 try:
                     queue_item = self.__invoker.services.queue.get()
@@ -49,6 +75,10 @@ class DefaultInvocationProcessor(InvocationProcessorABC):
                     # do not hammer the queue
                     time.sleep(0.5)
                     continue
+
+                if profiler and profiler.profile_id != queue_item.graph_execution_state_id:
+                    profiler.start(profile_id=queue_item.graph_execution_state_id)
+
                 try:
                     graph_execution_state = self.__invoker.services.graph_execution_manager.get(
                         queue_item.graph_execution_state_id
@@ -132,13 +162,12 @@ class DefaultInvocationProcessor(InvocationProcessorABC):
                             source_node_id=source_node_id,
                             result=outputs.model_dump(),
                         )
-                    self.__invoker.services.performance_statistics.log_stats()
 
                 except KeyboardInterrupt:
                     pass
 
                 except CanceledException:
-                    self.__invoker.services.performance_statistics.reset_stats(graph_execution_state.id)
+                    stats_cleanup(graph_execution_state.id)
                     pass
 
                 except Exception as e:
@@ -163,7 +192,6 @@ class DefaultInvocationProcessor(InvocationProcessorABC):
                         error_type=e.__class__.__name__,
                         error=error,
                     )
-                    self.__invoker.services.performance_statistics.reset_stats(graph_execution_state.id)
                     pass
 
                 # Check queue to see if this is canceled, and skip if so
@@ -201,6 +229,7 @@ class DefaultInvocationProcessor(InvocationProcessorABC):
                         queue_id=queue_item.session_queue_id,
                         graph_execution_state_id=graph_execution_state.id,
                     )
+                    stats_cleanup(graph_execution_state.id)
 
         except KeyboardInterrupt:
             pass  # Log something? KeyboardInterrupt is probably not going to be seen by the processor

invokeai/app/services/invocation_services.py

@@ -11,6 +11,7 @@ if TYPE_CHECKING:
     from .board_records.board_records_base import BoardRecordStorageBase
     from .boards.boards_base import BoardServiceABC
     from .config import InvokeAIAppConfig
+    from .download import DownloadQueueServiceBase
     from .events.events_base import EventServiceBase
     from .image_files.image_files_base import ImageFileStorageBase
     from .image_records.image_records_base import ImageRecordStorageBase
@@ -21,13 +22,11 @@ if TYPE_CHECKING:
     from .invocation_stats.invocation_stats_base import InvocationStatsServiceBase
     from .item_storage.item_storage_base import ItemStorageABC
     from .latents_storage.latents_storage_base import LatentsStorageBase
-    from .model_install import ModelInstallServiceBase
     from .model_manager.model_manager_base import ModelManagerServiceBase
-    from .model_records import ModelRecordServiceBase
     from .names.names_base import NameServiceBase
     from .session_processor.session_processor_base import SessionProcessorBase
     from .session_queue.session_queue_base import SessionQueueBase
-    from .shared.graph import GraphExecutionState, LibraryGraph
+    from .shared.graph import GraphExecutionState
     from .urls.urls_base import UrlServiceBase
     from .workflow_records.workflow_records_base import WorkflowRecordsStorageBase
 
@@ -43,15 +42,13 @@ class InvocationServices:
     configuration: "InvokeAIAppConfig"
     events: "EventServiceBase"
     graph_execution_manager: "ItemStorageABC[GraphExecutionState]"
-    graph_library: "ItemStorageABC[LibraryGraph]"
     images: "ImageServiceABC"
     image_records: "ImageRecordStorageBase"
     image_files: "ImageFileStorageBase"
     latents: "LatentsStorageBase"
     logger: "Logger"
     model_manager: "ModelManagerServiceBase"
-    model_records: "ModelRecordServiceBase"
-    model_install: "ModelInstallServiceBase"
+    download_queue: "DownloadQueueServiceBase"
     processor: "InvocationProcessorABC"
     performance_statistics: "InvocationStatsServiceBase"
     queue: "InvocationQueueABC"
@@ -71,15 +68,13 @@ class InvocationServices:
         configuration: "InvokeAIAppConfig",
         events: "EventServiceBase",
         graph_execution_manager: "ItemStorageABC[GraphExecutionState]",
-        graph_library: "ItemStorageABC[LibraryGraph]",
         images: "ImageServiceABC",
         image_files: "ImageFileStorageBase",
         image_records: "ImageRecordStorageBase",
         latents: "LatentsStorageBase",
         logger: "Logger",
         model_manager: "ModelManagerServiceBase",
-        model_records: "ModelRecordServiceBase",
-        model_install: "ModelInstallServiceBase",
+        download_queue: "DownloadQueueServiceBase",
         processor: "InvocationProcessorABC",
         performance_statistics: "InvocationStatsServiceBase",
         queue: "InvocationQueueABC",
@@ -97,15 +92,13 @@ class InvocationServices:
         self.configuration = configuration
         self.events = events
         self.graph_execution_manager = graph_execution_manager
-        self.graph_library = graph_library
         self.images = images
         self.image_files = image_files
         self.image_records = image_records
         self.latents = latents
         self.logger = logger
         self.model_manager = model_manager
-        self.model_records = model_records
-        self.model_install = model_install
+        self.download_queue = download_queue
         self.processor = processor
         self.performance_statistics = performance_statistics
         self.queue = queue

invokeai/app/services/invocation_stats/invocation_stats_base.py

@@ -29,37 +29,28 @@ writes to the system log is stored in InvocationServices.performance_statistics.
 """
 
 from abc import ABC, abstractmethod
-from contextlib import AbstractContextManager
-from typing import Dict
+from pathlib import Path
+from typing import Iterator
 
 from invokeai.app.invocations.baseinvocation import BaseInvocation
-from invokeai.backend.model_management.model_cache import CacheStats
-
-from .invocation_stats_common import NodeLog
+from invokeai.app.services.invocation_stats.invocation_stats_common import InvocationStatsSummary
 
 
 class InvocationStatsServiceBase(ABC):
     "Abstract base class for recording node memory/time performance statistics"
 
-    # {graph_id => NodeLog}
-    _stats: Dict[str, NodeLog]
-    _cache_stats: Dict[str, CacheStats]
-    ram_used: float
-    ram_changed: float
-
     @abstractmethod
-    def __init__(self):
+    def __init__(self) -> None:
         """
         Initialize the InvocationStatsService and reset counters to zero
         """
-        pass
 
     @abstractmethod
     def collect_stats(
         self,
         invocation: BaseInvocation,
         graph_execution_state_id: str,
-    ) -> AbstractContextManager:
+    ) -> Iterator[None]:
         """
         Return a context object that will capture the statistics on the execution
         of invocaation. Use with: to place around the part of the code that executes the invocation.
@@ -69,53 +60,38 @@ class InvocationStatsServiceBase(ABC):
         pass
 
     @abstractmethod
-    def reset_stats(self, graph_execution_state_id: str):
+    def reset_stats(self, graph_execution_state_id: str) -> None:
         """
-        Reset all statistics for the indicated graph
-        :param graph_execution_state_id
+        Reset all statistics for the indicated graph.
+        :param graph_execution_state_id: The id of the session whose stats to reset.
+        :raises GESStatsNotFoundError: if the graph isn't tracked in the stats.
         """
         pass
 
     @abstractmethod
-    def reset_all_stats(self):
-        """Zero all statistics"""
-        pass
-
-    @abstractmethod
-    def update_invocation_stats(
-        self,
-        graph_id: str,
-        invocation_type: str,
-        time_used: float,
-        vram_used: float,
-    ):
-        """
-        Add timing information on execution of a node. Usually
-        used internally.
-        :param graph_id: ID of the graph that is currently executing
-        :param invocation_type: String literal type of the node
-        :param time_used: Time used by node's exection (sec)
-        :param vram_used: Maximum VRAM used during exection (GB)
-        """
-        pass
-
-    @abstractmethod
-    def log_stats(self):
+    def log_stats(self, graph_execution_state_id: str) -> None:
         """
         Write out the accumulated statistics to the log or somewhere else.
+        :param graph_execution_state_id: The id of the session whose stats to log.
+        :raises GESStatsNotFoundError: if the graph isn't tracked in the stats.
         """
         pass
 
     @abstractmethod
-    def update_mem_stats(
-        self,
-        ram_used: float,
-        ram_changed: float,
-    ):
+    def get_stats(self, graph_execution_state_id: str) -> InvocationStatsSummary:
         """
-        Update the collector with RAM memory usage info.
-
-        :param ram_used: How much RAM is currently in use.
-        :param ram_changed: How much RAM changed since last generation.
+        Gets the accumulated statistics for the indicated graph.
+        :param graph_execution_state_id: The id of the session whose stats to get.
+        :raises GESStatsNotFoundError: if the graph isn't tracked in the stats.
+        """
+        pass
+
+    @abstractmethod
+    def dump_stats(self, graph_execution_state_id: str, output_path: Path) -> None:
+        """
+        Write out the accumulated statistics to the indicated path as JSON.
+        :param graph_execution_state_id: The id of the session whose stats to dump.
+        :param output_path: The file to write the stats to.
+        :raises GESStatsNotFoundError: if the graph isn't tracked in the stats.
         """
         pass

invokeai/app/services/invocation_stats/invocation_stats_common.py

@@ -1,25 +1,183 @@
-from dataclasses import dataclass, field
-from typing import Dict
+from collections import defaultdict
+from dataclasses import asdict, dataclass
+from typing import Any, Optional
 
-# size of GIG in bytes
-GIG = 1073741824
+
+class GESStatsNotFoundError(Exception):
+    """Raised when execution stats are not found for a given Graph Execution State."""
 
 
 @dataclass
-class NodeStats:
-    """Class for tracking execution stats of an invocation node"""
+class NodeExecutionStatsSummary:
+    """The stats for a specific type of node."""
 
-    calls: int = 0
-    time_used: float = 0.0  # seconds
-    max_vram: float = 0.0  # GB
-    cache_hits: int = 0
-    cache_misses: int = 0
-    cache_high_watermark: int = 0
+    node_type: str
+    num_calls: int
+    time_used_seconds: float
+    peak_vram_gb: float
 
 
 @dataclass
-class NodeLog:
-    """Class for tracking node usage"""
+class ModelCacheStatsSummary:
+    """The stats for the model cache."""
 
-    # {node_type => NodeStats}
-    nodes: Dict[str, NodeStats] = field(default_factory=dict)
+    high_water_mark_gb: float
+    cache_size_gb: float
+    total_usage_gb: float
+    cache_hits: int
+    cache_misses: int
+    models_cached: int
+    models_cleared: int
+
+
+@dataclass
+class GraphExecutionStatsSummary:
+    """The stats for the graph execution state."""
+
+    graph_execution_state_id: str
+    execution_time_seconds: float
+    # `wall_time_seconds`, `ram_usage_gb` and `ram_change_gb` are derived from the node execution stats.
+    # In some situations, there are no node stats, so these values are optional.
+    wall_time_seconds: Optional[float]
+    ram_usage_gb: Optional[float]
+    ram_change_gb: Optional[float]
+
+
+@dataclass
+class InvocationStatsSummary:
+    """
+    The accumulated stats for a graph execution.
+    Its `__str__` method returns a human-readable stats summary.
+    """
+
+    vram_usage_gb: Optional[float]
+    graph_stats: GraphExecutionStatsSummary
+    model_cache_stats: ModelCacheStatsSummary
+    node_stats: list[NodeExecutionStatsSummary]
+
+    def __str__(self) -> str:
+        _str = ""
+        _str = f"Graph stats: {self.graph_stats.graph_execution_state_id}\n"
+        _str += f"{'Node':>30} {'Calls':>7} {'Seconds':>9} {'VRAM Used':>10}\n"
+
+        for summary in self.node_stats:
+            _str += f"{summary.node_type:>30} {summary.num_calls:>7} {summary.time_used_seconds:>8.3f}s {summary.peak_vram_gb:>9.3f}G\n"
+
+        _str += f"TOTAL GRAPH EXECUTION TIME: {self.graph_stats.execution_time_seconds:7.3f}s\n"
+
+        if self.graph_stats.wall_time_seconds is not None:
+            _str += f"TOTAL GRAPH WALL TIME: {self.graph_stats.wall_time_seconds:7.3f}s\n"
+
+        if self.graph_stats.ram_usage_gb is not None and self.graph_stats.ram_change_gb is not None:
+            _str += f"RAM used by InvokeAI process: {self.graph_stats.ram_usage_gb:4.2f}G ({self.graph_stats.ram_change_gb:+5.3f}G)\n"
+
+        _str += f"RAM used to load models: {self.model_cache_stats.total_usage_gb:4.2f}G\n"
+        if self.vram_usage_gb:
+            _str += f"VRAM in use: {self.vram_usage_gb:4.3f}G\n"
+        _str += "RAM cache statistics:\n"
+        _str += f"   Model cache hits: {self.model_cache_stats.cache_hits}\n"
+        _str += f"   Model cache misses: {self.model_cache_stats.cache_misses}\n"
+        _str += f"   Models cached: {self.model_cache_stats.models_cached}\n"
+        _str += f"   Models cleared from cache: {self.model_cache_stats.models_cleared}\n"
+        _str += f"   Cache high water mark: {self.model_cache_stats.high_water_mark_gb:4.2f}/{self.model_cache_stats.cache_size_gb:4.2f}G\n"
+
+        return _str
+
+    def as_dict(self) -> dict[str, Any]:
+        """Returns the stats as a dictionary."""
+        return asdict(self)
+
+
+@dataclass
+class NodeExecutionStats:
+    """Class for tracking execution stats of an invocation node."""
+
+    invocation_type: str
+
+    start_time: float  # Seconds since the epoch.
+    end_time: float  # Seconds since the epoch.
+
+    start_ram_gb: float  # GB
+    end_ram_gb: float  # GB
+
+    peak_vram_gb: float  # GB
+
+    def total_time(self) -> float:
+        return self.end_time - self.start_time
+
+
+class GraphExecutionStats:
+    """Class for tracking execution stats of a graph."""
+
+    def __init__(self):
+        self._node_stats_list: list[NodeExecutionStats] = []
+
+    def add_node_execution_stats(self, node_stats: NodeExecutionStats):
+        self._node_stats_list.append(node_stats)
+
+    def get_total_run_time(self) -> float:
+        """Get the total time spent executing nodes in the graph."""
+        total = 0.0
+        for node_stats in self._node_stats_list:
+            total += node_stats.total_time()
+        return total
+
+    def get_first_node_stats(self) -> NodeExecutionStats | None:
+        """Get the stats of the first node in the graph (by start_time)."""
+        first_node = None
+        for node_stats in self._node_stats_list:
+            if first_node is None or node_stats.start_time < first_node.start_time:
+                first_node = node_stats
+
+        assert first_node is not None
+        return first_node
+
+    def get_last_node_stats(self) -> NodeExecutionStats | None:
+        """Get the stats of the last node in the graph (by end_time)."""
+        last_node = None
+        for node_stats in self._node_stats_list:
+            if last_node is None or node_stats.end_time > last_node.end_time:
+                last_node = node_stats
+
+        return last_node
+
+    def get_graph_stats_summary(self, graph_execution_state_id: str) -> GraphExecutionStatsSummary:
+        """Get a summary of the graph stats."""
+        first_node = self.get_first_node_stats()
+        last_node = self.get_last_node_stats()
+
+        wall_time_seconds: Optional[float] = None
+        ram_usage_gb: Optional[float] = None
+        ram_change_gb: Optional[float] = None
+
+        if last_node and first_node:
+            wall_time_seconds = last_node.end_time - first_node.start_time
+            ram_usage_gb = last_node.end_ram_gb
+            ram_change_gb = last_node.end_ram_gb - first_node.start_ram_gb
+
+        return GraphExecutionStatsSummary(
+            graph_execution_state_id=graph_execution_state_id,
+            execution_time_seconds=self.get_total_run_time(),
+            wall_time_seconds=wall_time_seconds,
+            ram_usage_gb=ram_usage_gb,
+            ram_change_gb=ram_change_gb,
+        )
+
+    def get_node_stats_summaries(self) -> list[NodeExecutionStatsSummary]:
+        """Get a summary of the node stats."""
+        summaries: list[NodeExecutionStatsSummary] = []
+        node_stats_by_type: dict[str, list[NodeExecutionStats]] = defaultdict(list)
+
+        for node_stats in self._node_stats_list:
+            node_stats_by_type[node_stats.invocation_type].append(node_stats)
+
+        for node_type, node_type_stats_list in node_stats_by_type.items():
+            num_calls = len(node_type_stats_list)
+            time_used = sum([n.total_time() for n in node_type_stats_list])
+            peak_vram = max([n.peak_vram_gb for n in node_type_stats_list])
+            summary = NodeExecutionStatsSummary(
+                node_type=node_type, num_calls=num_calls, time_used_seconds=time_used, peak_vram_gb=peak_vram
+            )
+            summaries.append(summary)
+
+        return summaries

invokeai/app/services/invocation_stats/invocation_stats_default.py

@@ -1,5 +1,8 @@
+import json
 import time
-from typing import Dict
+from contextlib import contextmanager
+from pathlib import Path
+from typing import Iterator
 
 import psutil
 import torch
@@ -7,161 +10,167 @@ import torch
 import invokeai.backend.util.logging as logger
 from invokeai.app.invocations.baseinvocation import BaseInvocation
 from invokeai.app.services.invoker import Invoker
-from invokeai.app.services.model_manager.model_manager_base import ModelManagerServiceBase
-from invokeai.backend.model_management.model_cache import CacheStats
+from invokeai.app.services.item_storage.item_storage_common import ItemNotFoundError
+from invokeai.backend.model_manager.load.model_cache import CacheStats
 
 from .invocation_stats_base import InvocationStatsServiceBase
-from .invocation_stats_common import GIG, NodeLog, NodeStats
+from .invocation_stats_common import (
+    GESStatsNotFoundError,
+    GraphExecutionStats,
+    GraphExecutionStatsSummary,
+    InvocationStatsSummary,
+    ModelCacheStatsSummary,
+    NodeExecutionStats,
+    NodeExecutionStatsSummary,
+)
+
+# Size of 1GB in bytes.
+GB = 2**30
 
 
 class InvocationStatsService(InvocationStatsServiceBase):
     """Accumulate performance information about a running graph. Collects time spent in each node,
     as well as the maximum and current VRAM utilisation for CUDA systems"""
 
-    _invoker: Invoker
-
     def __init__(self):
-        # {graph_id => NodeLog}
-        self._stats: Dict[str, NodeLog] = {}
-        self._cache_stats: Dict[str, CacheStats] = {}
-        self.ram_used: float = 0.0
-        self.ram_changed: float = 0.0
+        # Maps graph_execution_state_id to GraphExecutionStats.
+        self._stats: dict[str, GraphExecutionStats] = {}
+        # Maps graph_execution_state_id to model manager CacheStats.
+        self._cache_stats: dict[str, CacheStats] = {}
 
     def start(self, invoker: Invoker) -> None:
         self._invoker = invoker
 
-    class StatsContext:
-        """Context manager for collecting statistics."""
-
-        invocation: BaseInvocation
-        collector: "InvocationStatsServiceBase"
-        graph_id: str
-        start_time: float
-        ram_used: int
-        model_manager: ModelManagerServiceBase
-
-        def __init__(
-            self,
-            invocation: BaseInvocation,
-            graph_id: str,
-            model_manager: ModelManagerServiceBase,
-            collector: "InvocationStatsServiceBase",
-        ):
-            """Initialize statistics for this run."""
-            self.invocation = invocation
-            self.collector = collector
-            self.graph_id = graph_id
-            self.start_time = 0.0
-            self.ram_used = 0
-            self.model_manager = model_manager
-
-        def __enter__(self):
-            self.start_time = time.time()
-            if torch.cuda.is_available():
-                torch.cuda.reset_peak_memory_stats()
-            self.ram_used = psutil.Process().memory_info().rss
-            if self.model_manager:
-                self.model_manager.collect_cache_stats(self.collector._cache_stats[self.graph_id])
-
-        def __exit__(self, *args):
-            """Called on exit from the context."""
-            ram_used = psutil.Process().memory_info().rss
-            self.collector.update_mem_stats(
-                ram_used=ram_used / GIG,
-                ram_changed=(ram_used - self.ram_used) / GIG,
-            )
-            self.collector.update_invocation_stats(
-                graph_id=self.graph_id,
-                invocation_type=self.invocation.type,  # type: ignore # `type` is not on the `BaseInvocation` model, but *is* on all invocations
-                time_used=time.time() - self.start_time,
-                vram_used=torch.cuda.max_memory_allocated() / GIG if torch.cuda.is_available() else 0.0,
-            )
-
-    def collect_stats(
-        self,
-        invocation: BaseInvocation,
-        graph_execution_state_id: str,
-    ) -> StatsContext:
-        if not self._stats.get(graph_execution_state_id):  # first time we're seeing this
-            self._stats[graph_execution_state_id] = NodeLog()
+    @contextmanager
+    def collect_stats(self, invocation: BaseInvocation, graph_execution_state_id: str) -> Iterator[None]:
+        # This is to handle case of the model manager not being initialized, which happens
+        # during some tests.
+        services = self._invoker.services
+        if not self._stats.get(graph_execution_state_id):
+            # First time we're seeing this graph_execution_state_id.
+            self._stats[graph_execution_state_id] = GraphExecutionStats()
             self._cache_stats[graph_execution_state_id] = CacheStats()
-        return self.StatsContext(invocation, graph_execution_state_id, self._invoker.services.model_manager, self)
 
-    def reset_all_stats(self):
-        """Zero all statistics"""
-        self._stats = {}
+            # Prune stale stats. There should be none since we're starting a new graph, but just in case.
+            self._prune_stale_stats()
+
+        # Record state before the invocation.
+        start_time = time.time()
+        start_ram = psutil.Process().memory_info().rss
+        if torch.cuda.is_available():
+            torch.cuda.reset_peak_memory_stats()
+
+        assert services.model_manager.load is not None
+        services.model_manager.load.ram_cache.stats = self._cache_stats[graph_execution_state_id]
 
-    def reset_stats(self, graph_execution_id: str):
         try:
-            self._stats.pop(graph_execution_id)
-        except KeyError:
-            logger.warning(f"Attempted to clear statistics for unknown graph {graph_execution_id}")
+            # Let the invocation run.
+            yield None
+        finally:
+            # Record state after the invocation.
+            node_stats = NodeExecutionStats(
+                invocation_type=invocation.get_type(),
+                start_time=start_time,
+                end_time=time.time(),
+                start_ram_gb=start_ram / GB,
+                end_ram_gb=psutil.Process().memory_info().rss / GB,
+                peak_vram_gb=torch.cuda.max_memory_allocated() / GB if torch.cuda.is_available() else 0.0,
+            )
+            self._stats[graph_execution_state_id].add_node_execution_stats(node_stats)
 
-    def update_mem_stats(
-        self,
-        ram_used: float,
-        ram_changed: float,
-    ):
-        self.ram_used = ram_used
-        self.ram_changed = ram_changed
+    def _prune_stale_stats(self) -> None:
+        """Check all graphs being tracked and prune any that have completed/errored.
 
-    def update_invocation_stats(
-        self,
-        graph_id: str,
-        invocation_type: str,
-        time_used: float,
-        vram_used: float,
-    ):
-        if not self._stats[graph_id].nodes.get(invocation_type):
-            self._stats[graph_id].nodes[invocation_type] = NodeStats()
-        stats = self._stats[graph_id].nodes[invocation_type]
-        stats.calls += 1
-        stats.time_used += time_used
-        stats.max_vram = max(stats.max_vram, vram_used)
-
-    def log_stats(self):
-        completed = set()
-        errored = set()
-        for graph_id, _node_log in self._stats.items():
+        This shouldn't be necessary, but we don't have totally robust upstream handling of graph completions/errors, so
+        for now we call this function periodically to prevent them from accumulating.
+        """
+        to_prune: list[str] = []
+        for graph_execution_state_id in self._stats:
             try:
-                current_graph_state = self._invoker.services.graph_execution_manager.get(graph_id)
-            except Exception:
-                errored.add(graph_id)
+                graph_execution_state = self._invoker.services.graph_execution_manager.get(graph_execution_state_id)
+            except ItemNotFoundError:
+                # TODO(ryand): What would cause this? Should this exception just be allowed to propagate?
+                logger.warning(f"Failed to get graph state for {graph_execution_state_id}.")
                 continue
 
-            if not current_graph_state.is_complete():
+            if not graph_execution_state.is_complete():
+                # The graph is still running, don't prune it.
                 continue
 
-            total_time = 0
-            logger.info(f"Graph stats: {graph_id}")
-            logger.info(f"{'Node':>30} {'Calls':>7}{'Seconds':>9} {'VRAM Used':>10}")
-            for node_type, stats in self._stats[graph_id].nodes.items():
-                logger.info(f"{node_type:>30}  {stats.calls:>4}   {stats.time_used:7.3f}s     {stats.max_vram:4.3f}G")
-                total_time += stats.time_used
+            to_prune.append(graph_execution_state_id)
 
-            cache_stats = self._cache_stats[graph_id]
-            hwm = cache_stats.high_watermark / GIG
-            tot = cache_stats.cache_size / GIG
-            loaded = sum(list(cache_stats.loaded_model_sizes.values())) / GIG
+        for graph_execution_state_id in to_prune:
+            del self._stats[graph_execution_state_id]
+            del self._cache_stats[graph_execution_state_id]
 
-            logger.info(f"TOTAL GRAPH EXECUTION TIME:  {total_time:7.3f}s")
-            logger.info("RAM used by InvokeAI process: " + "%4.2fG" % self.ram_used + f" ({self.ram_changed:+5.3f}G)")
-            logger.info(f"RAM used to load models: {loaded:4.2f}G")
-            if torch.cuda.is_available():
-                logger.info("VRAM in use: " + "%4.3fG" % (torch.cuda.memory_allocated() / GIG))
-            logger.info("RAM cache statistics:")
-            logger.info(f"   Model cache hits: {cache_stats.hits}")
-            logger.info(f"   Model cache misses: {cache_stats.misses}")
-            logger.info(f"   Models cached: {cache_stats.in_cache}")
-            logger.info(f"   Models cleared from cache: {cache_stats.cleared}")
-            logger.info(f"   Cache high water mark: {hwm:4.2f}/{tot:4.2f}G")
+        if len(to_prune) > 0:
+            logger.info(f"Pruned stale graph stats for {to_prune}.")
 
-            completed.add(graph_id)
+    def reset_stats(self, graph_execution_state_id: str):
+        try:
+            del self._stats[graph_execution_state_id]
+            del self._cache_stats[graph_execution_state_id]
+        except KeyError as e:
+            raise GESStatsNotFoundError(
+                f"Attempted to clear statistics for unknown graph {graph_execution_state_id}: {e}."
+            ) from e
 
-        for graph_id in completed:
-            del self._stats[graph_id]
-            del self._cache_stats[graph_id]
+    def get_stats(self, graph_execution_state_id: str) -> InvocationStatsSummary:
+        graph_stats_summary = self._get_graph_summary(graph_execution_state_id)
+        node_stats_summaries = self._get_node_summaries(graph_execution_state_id)
+        model_cache_stats_summary = self._get_model_cache_summary(graph_execution_state_id)
+        vram_usage_gb = torch.cuda.memory_allocated() / GB if torch.cuda.is_available() else None
 
-        for graph_id in errored:
-            del self._stats[graph_id]
-            del self._cache_stats[graph_id]
+        return InvocationStatsSummary(
+            graph_stats=graph_stats_summary,
+            model_cache_stats=model_cache_stats_summary,
+            node_stats=node_stats_summaries,
+            vram_usage_gb=vram_usage_gb,
+        )
+
+    def log_stats(self, graph_execution_state_id: str) -> None:
+        stats = self.get_stats(graph_execution_state_id)
+        logger.info(str(stats))
+
+    def dump_stats(self, graph_execution_state_id: str, output_path: Path) -> None:
+        stats = self.get_stats(graph_execution_state_id)
+        with open(output_path, "w") as f:
+            f.write(json.dumps(stats.as_dict(), indent=2))
+
+    def _get_model_cache_summary(self, graph_execution_state_id: str) -> ModelCacheStatsSummary:
+        try:
+            cache_stats = self._cache_stats[graph_execution_state_id]
+        except KeyError as e:
+            raise GESStatsNotFoundError(
+                f"Attempted to get model cache statistics for unknown graph {graph_execution_state_id}: {e}."
+            ) from e
+
+        return ModelCacheStatsSummary(
+            cache_hits=cache_stats.hits,
+            cache_misses=cache_stats.misses,
+            high_water_mark_gb=cache_stats.high_watermark / GB,
+            cache_size_gb=cache_stats.cache_size / GB,
+            total_usage_gb=sum(list(cache_stats.loaded_model_sizes.values())) / GB,
+            models_cached=cache_stats.in_cache,
+            models_cleared=cache_stats.cleared,
+        )
+
+    def _get_graph_summary(self, graph_execution_state_id: str) -> GraphExecutionStatsSummary:
+        try:
+            graph_stats = self._stats[graph_execution_state_id]
+        except KeyError as e:
+            raise GESStatsNotFoundError(
+                f"Attempted to get graph statistics for unknown graph {graph_execution_state_id}: {e}."
+            ) from e
+
+        return graph_stats.get_graph_stats_summary(graph_execution_state_id)
+
+    def _get_node_summaries(self, graph_execution_state_id: str) -> list[NodeExecutionStatsSummary]:
+        try:
+            graph_stats = self._stats[graph_execution_state_id]
+        except KeyError as e:
+            raise GESStatsNotFoundError(
+                f"Attempted to get node statistics for unknown graph {graph_execution_state_id}: {e}."
+            ) from e
+
+        return graph_stats.get_node_stats_summaries()

invokeai/app/services/item_storage/item_storage_base.py

@@ -1,10 +1,8 @@
 from abc import ABC, abstractmethod
-from typing import Callable, Generic, Optional, TypeVar
+from typing import Callable, Generic, TypeVar
 
 from pydantic import BaseModel
 
-from invokeai.app.services.shared.pagination import PaginatedResults
-
 T = TypeVar("T", bound=BaseModel)
 
 
@@ -22,26 +20,26 @@ class ItemStorageABC(ABC, Generic[T]):
 
     @abstractmethod
     def get(self, item_id: str) -> T:
-        """Gets the item, parsing it into a Pydantic model"""
-        pass
-
-    @abstractmethod
-    def get_raw(self, item_id: str) -> Optional[str]:
-        """Gets the raw item as a string, skipping Pydantic parsing"""
+        """
+        Gets the item.
+        :param item_id: the id of the item to get
+        :raises ItemNotFoundError: if the item is not found
+        """
         pass
 
     @abstractmethod
     def set(self, item: T) -> None:
-        """Sets the item"""
+        """
+        Sets the item. The id will be extracted based on id_field.
+        :param item: the item to set
+        """
         pass
 
     @abstractmethod
-    def list(self, page: int = 0, per_page: int = 10) -> PaginatedResults[T]:
-        """Gets a paginated list of items"""
-        pass
-
-    @abstractmethod
-    def search(self, query: str, page: int = 0, per_page: int = 10) -> PaginatedResults[T]:
+    def delete(self, item_id: str) -> None:
+        """
+        Deletes the item, if it exists.
+        """
         pass
 
     def on_changed(self, on_changed: Callable[[T], None]) -> None:

invokeai/app/services/item_storage/item_storage_common.py

@@ -0,0 +1,5 @@
+class ItemNotFoundError(KeyError):
+    """Raised when an item is not found in storage"""
+
+    def __init__(self, item_id: str) -> None:
+        super().__init__(f"Item with id {item_id} not found")

invokeai/app/services/item_storage/item_storage_memory.py

@@ -0,0 +1,52 @@
+from collections import OrderedDict
+from contextlib import suppress
+from typing import Generic, TypeVar
+
+from pydantic import BaseModel
+
+from invokeai.app.services.item_storage.item_storage_base import ItemStorageABC
+from invokeai.app.services.item_storage.item_storage_common import ItemNotFoundError
+
+T = TypeVar("T", bound=BaseModel)
+
+
+class ItemStorageMemory(ItemStorageABC[T], Generic[T]):
+    """
+    Provides a simple in-memory storage for items, with a maximum number of items to store.
+    The storage uses the LRU strategy to evict items from storage when the max has been reached.
+    """
+
+    def __init__(self, id_field: str = "id", max_items: int = 10) -> None:
+        super().__init__()
+        if max_items < 1:
+            raise ValueError("max_items must be at least 1")
+        if not id_field:
+            raise ValueError("id_field must not be empty")
+        self._id_field = id_field
+        self._items: OrderedDict[str, T] = OrderedDict()
+        self._max_items = max_items
+
+    def get(self, item_id: str) -> T:
+        # If the item exists, move it to the end of the OrderedDict.
+        item = self._items.pop(item_id, None)
+        if item is None:
+            raise ItemNotFoundError(item_id)
+        self._items[item_id] = item
+        return item
+
+    def set(self, item: T) -> None:
+        item_id = getattr(item, self._id_field)
+        if item_id in self._items:
+            # If item already exists, remove it and add it to the end
+            self._items.pop(item_id)
+        elif len(self._items) >= self._max_items:
+            # If cache is full, evict the least recently used item
+            self._items.popitem(last=False)
+        self._items[item_id] = item
+        self._on_changed(item)
+
+    def delete(self, item_id: str) -> None:
+        # This is a no-op if the item doesn't exist.
+        with suppress(KeyError):
+            del self._items[item_id]
+            self._on_deleted(item_id)

invokeai/app/services/item_storage/item_storage_sqlite.py

@@ -1,147 +0,0 @@
-import sqlite3
-import threading
-from typing import Generic, Optional, TypeVar, get_args
-
-from pydantic import BaseModel, TypeAdapter
-
-from invokeai.app.services.shared.pagination import PaginatedResults
-from invokeai.app.services.shared.sqlite.sqlite_database import SqliteDatabase
-
-from .item_storage_base import ItemStorageABC
-
-T = TypeVar("T", bound=BaseModel)
-
-
-class SqliteItemStorage(ItemStorageABC, Generic[T]):
-    _table_name: str
-    _conn: sqlite3.Connection
-    _cursor: sqlite3.Cursor
-    _id_field: str
-    _lock: threading.RLock
-    _validator: Optional[TypeAdapter[T]]
-
-    def __init__(self, db: SqliteDatabase, table_name: str, id_field: str = "id"):
-        super().__init__()
-
-        self._lock = db.lock
-        self._conn = db.conn
-        self._table_name = table_name
-        self._id_field = id_field  # TODO: validate that T has this field
-        self._cursor = self._conn.cursor()
-        self._validator: Optional[TypeAdapter[T]] = None
-
-        self._create_table()
-
-    def _create_table(self):
-        try:
-            self._lock.acquire()
-            self._cursor.execute(
-                f"""CREATE TABLE IF NOT EXISTS {self._table_name} (
-                item TEXT,
-                id TEXT GENERATED ALWAYS AS (json_extract(item, '$.{self._id_field}')) VIRTUAL NOT NULL);"""
-            )
-            self._cursor.execute(
-                f"""CREATE UNIQUE INDEX IF NOT EXISTS {self._table_name}_id ON {self._table_name}(id);"""
-            )
-        finally:
-            self._lock.release()
-
-    def _parse_item(self, item: str) -> T:
-        if self._validator is None:
-            """
-            We don't get access to `__orig_class__` in `__init__()`, and we need this before start(), so
-            we can create it when it is first needed instead.
-            __orig_class__ is technically an implementation detail of the typing module, not a supported API
-            """
-            self._validator = TypeAdapter(get_args(self.__orig_class__)[0])  # type: ignore [attr-defined]
-        return self._validator.validate_json(item)
-
-    def set(self, item: T):
-        try:
-            self._lock.acquire()
-            self._cursor.execute(
-                f"""INSERT OR REPLACE INTO {self._table_name} (item) VALUES (?);""",
-                (item.model_dump_json(warnings=False, exclude_none=True),),
-            )
-            self._conn.commit()
-        finally:
-            self._lock.release()
-        self._on_changed(item)
-
-    def get(self, id: str) -> Optional[T]:
-        try:
-            self._lock.acquire()
-            self._cursor.execute(f"""SELECT item FROM {self._table_name} WHERE id = ?;""", (str(id),))
-            result = self._cursor.fetchone()
-        finally:
-            self._lock.release()
-
-        if not result:
-            return None
-
-        return self._parse_item(result[0])
-
-    def get_raw(self, id: str) -> Optional[str]:
-        try:
-            self._lock.acquire()
-            self._cursor.execute(f"""SELECT item FROM {self._table_name} WHERE id = ?;""", (str(id),))
-            result = self._cursor.fetchone()
-        finally:
-            self._lock.release()
-
-        if not result:
-            return None
-
-        return result[0]
-
-    def delete(self, id: str):
-        try:
-            self._lock.acquire()
-            self._cursor.execute(f"""DELETE FROM {self._table_name} WHERE id = ?;""", (str(id),))
-            self._conn.commit()
-        finally:
-            self._lock.release()
-        self._on_deleted(id)
-
-    def list(self, page: int = 0, per_page: int = 10) -> PaginatedResults[T]:
-        try:
-            self._lock.acquire()
-            self._cursor.execute(
-                f"""SELECT item FROM {self._table_name} LIMIT ? OFFSET ?;""",
-                (per_page, page * per_page),
-            )
-            result = self._cursor.fetchall()
-
-            items = [self._parse_item(r[0]) for r in result]
-
-            self._cursor.execute(f"""SELECT count(*) FROM {self._table_name};""")
-            count = self._cursor.fetchone()[0]
-        finally:
-            self._lock.release()
-
-        pageCount = int(count / per_page) + 1
-
-        return PaginatedResults[T](items=items, page=page, pages=pageCount, per_page=per_page, total=count)
-
-    def search(self, query: str, page: int = 0, per_page: int = 10) -> PaginatedResults[T]:
-        try:
-            self._lock.acquire()
-            self._cursor.execute(
-                f"""SELECT item FROM {self._table_name} WHERE item LIKE ? LIMIT ? OFFSET ?;""",
-                (f"%{query}%", per_page, page * per_page),
-            )
-            result = self._cursor.fetchall()
-
-            items = [self._parse_item(r[0]) for r in result]
-
-            self._cursor.execute(
-                f"""SELECT count(*) FROM {self._table_name} WHERE item LIKE ?;""",
-                (f"%{query}%",),
-            )
-            count = self._cursor.fetchone()[0]
-        finally:
-            self._lock.release()
-
-        pageCount = int(count / per_page) + 1
-
-        return PaginatedResults[T](items=items, page=page, pages=pageCount, per_page=per_page, total=count)