Mirrored_Repos/InvokeAI
1
0
Fork 0
mirror of https://github.com/invoke-ai/InvokeAI synced 2024-08-30 20:32:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'main' into stalker7779/backend_base

Browse Source
This commit is contained in:
StAlKeR7779 2024-07-18 01:08:04 +03:00 committed by GitHub
commit 710dc6b487
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
48 changed files with 4959 additions and 1999 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
invokeai/app/api/routers/images.py
View File

@ -233,21 +233,14 @@ async def get_image_workflow(
) )
async def get_image_full( async def get_image_full(
image_name: str = Path(description="The name of full-resolution image file to get"), image_name: str = Path(description="The name of full-resolution image file to get"),
) -> FileResponse: ) -> Response:
"""Gets a full-resolution image file""" """Gets a full-resolution image file"""
try: try:
path = ApiDependencies.invoker.services.images.get_path(image_name) path = ApiDependencies.invoker.services.images.get_path(image_name)
with open(path, "rb") as f:
if not ApiDependencies.invoker.services.images.validate_path(path): content = f.read()
raise HTTPException(status_code=404) response = Response(content, media_type="image/png")
response = FileResponse(
path,
media_type="image/png",
filename=image_name,
content_disposition_type="inline",
)
response.headers["Cache-Control"] = f"max-age={IMAGE_MAX_AGE}" response.headers["Cache-Control"] = f"max-age={IMAGE_MAX_AGE}"
return response return response
except Exception: except Exception:
@ -268,15 +261,14 @@ async def get_image_full(
) )
async def get_image_thumbnail( async def get_image_thumbnail(
image_name: str = Path(description="The name of thumbnail image file to get"), image_name: str = Path(description="The name of thumbnail image file to get"),
) -> FileResponse: ) -> Response:
"""Gets a thumbnail image file""" """Gets a thumbnail image file"""
try: try:
path = ApiDependencies.invoker.services.images.get_path(image_name, thumbnail=True) path = ApiDependencies.invoker.services.images.get_path(image_name, thumbnail=True)
if not ApiDependencies.invoker.services.images.validate_path(path): with open(path, "rb") as f:
raise HTTPException(status_code=404) content = f.read()
response = Response(content, media_type="image/webp")
response = FileResponse(path, media_type="image/webp", content_disposition_type="inline")
response.headers["Cache-Control"] = f"max-age={IMAGE_MAX_AGE}" response.headers["Cache-Control"] = f"max-age={IMAGE_MAX_AGE}"
return response return response
except Exception: except Exception:

1
invokeai/app/api_app.py
View File

@ -161,6 +161,7 @@ def invoke_api() -> None:
# Taken from https://waylonwalker.com/python-find-available-port/, thanks Waylon! # Taken from https://waylonwalker.com/python-find-available-port/, thanks Waylon!
# https://github.com/WaylonWalker # https://github.com/WaylonWalker
with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s: with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s:
s.settimeout(1)
if s.connect_ex(("localhost", port)) == 0: if s.connect_ex(("localhost", port)) == 0:
return find_port(port=port + 1) return find_port(port=port + 1)
else: else:

2
invokeai/app/invocations/fields.py
View File

@ -48,6 +48,7 @@ class UIType(str, Enum, metaclass=MetaEnum):
ControlNetModel = "ControlNetModelField" ControlNetModel = "ControlNetModelField"
IPAdapterModel = "IPAdapterModelField" IPAdapterModel = "IPAdapterModelField"
T2IAdapterModel = "T2IAdapterModelField" T2IAdapterModel = "T2IAdapterModelField"
SpandrelImageToImageModel = "SpandrelImageToImageModelField"
# endregion # endregion
# region Misc Field Types # region Misc Field Types
@ -134,6 +135,7 @@ class FieldDescriptions:
sdxl_main_model = "SDXL Main model (UNet, VAE, CLIP1, CLIP2) to load" sdxl_main_model = "SDXL Main model (UNet, VAE, CLIP1, CLIP2) to load"
sdxl_refiner_model = "SDXL Refiner Main Modde (UNet, VAE, CLIP2) to load" sdxl_refiner_model = "SDXL Refiner Main Modde (UNet, VAE, CLIP2) to load"
onnx_main_model = "ONNX Main model (UNet, VAE, CLIP) to load" onnx_main_model = "ONNX Main model (UNet, VAE, CLIP) to load"
spandrel_image_to_image_model = "Image-to-Image model"
lora_weight = "The weight at which the LoRA is applied to each model" lora_weight = "The weight at which the LoRA is applied to each model"
compel_prompt = "Prompt to be parsed by Compel to create a conditioning tensor" compel_prompt = "Prompt to be parsed by Compel to create a conditioning tensor"
raw_prompt = "Raw prompt text (no parsing)" raw_prompt = "Raw prompt text (no parsing)"

144
invokeai/app/invocations/spandrel_image_to_image.py Normal file
View File

@ -0,0 +1,144 @@
import numpy as np
import torch
from PIL import Image
from tqdm import tqdm
from invokeai.app.invocations.baseinvocation import BaseInvocation, invocation
from invokeai.app.invocations.fields import (
FieldDescriptions,
ImageField,
InputField,
UIType,
WithBoard,
WithMetadata,
)
from invokeai.app.invocations.model import ModelIdentifierField
from invokeai.app.invocations.primitives import ImageOutput
from invokeai.app.services.session_processor.session_processor_common import CanceledException
from invokeai.app.services.shared.invocation_context import InvocationContext
from invokeai.backend.spandrel_image_to_image_model import SpandrelImageToImageModel
from invokeai.backend.tiles.tiles import calc_tiles_min_overlap
from invokeai.backend.tiles.utils import TBLR, Tile
@invocation("spandrel_image_to_image", title="Image-to-Image", tags=["upscale"], category="upscale", version="1.1.0")
class SpandrelImageToImageInvocation(BaseInvocation, WithMetadata, WithBoard):
"""Run any spandrel image-to-image model (https://github.com/chaiNNer-org/spandrel)."""
image: ImageField = InputField(description="The input image")
image_to_image_model: ModelIdentifierField = InputField(
title="Image-to-Image Model",
description=FieldDescriptions.spandrel_image_to_image_model,
ui_type=UIType.SpandrelImageToImageModel,
)
tile_size: int = InputField(
default=512, description="The tile size for tiled image-to-image. Set to 0 to disable tiling."
)
def _scale_tile(self, tile: Tile, scale: int) -> Tile:
return Tile(
coords=TBLR(
top=tile.coords.top * scale,
bottom=tile.coords.bottom * scale,
left=tile.coords.left * scale,
right=tile.coords.right * scale,
),
overlap=TBLR(
top=tile.overlap.top * scale,
bottom=tile.overlap.bottom * scale,
left=tile.overlap.left * scale,
right=tile.overlap.right * scale,
),
)
@torch.inference_mode()
def invoke(self, context: InvocationContext) -> ImageOutput:
# Images are converted to RGB, because most models don't support an alpha channel. In the future, we may want to
# revisit this.
image = context.images.get_pil(self.image.image_name, mode="RGB")
# Compute the image tiles.
if self.tile_size > 0:
min_overlap = 20
tiles = calc_tiles_min_overlap(
image_height=image.height,
image_width=image.width,
tile_height=self.tile_size,
tile_width=self.tile_size,
min_overlap=min_overlap,
)
else:
# No tiling. Generate a single tile that covers the entire image.
min_overlap = 0
tiles = [
Tile(
coords=TBLR(top=0, bottom=image.height, left=0, right=image.width),
overlap=TBLR(top=0, bottom=0, left=0, right=0),
)
]
# Sort tiles first by left x coordinate, then by top y coordinate. During tile processing, we want to iterate
# over tiles left-to-right, top-to-bottom.
tiles = sorted(tiles, key=lambda x: x.coords.left)
tiles = sorted(tiles, key=lambda x: x.coords.top)
# Prepare input image for inference.
image_tensor = SpandrelImageToImageModel.pil_to_tensor(image)
# Load the model.
spandrel_model_info = context.models.load(self.image_to_image_model)
# Run the model on each tile.
with spandrel_model_info as spandrel_model:
assert isinstance(spandrel_model, SpandrelImageToImageModel)
# Scale the tiles for re-assembling the final image.
scale = spandrel_model.scale
scaled_tiles = [self._scale_tile(tile, scale=scale) for tile in tiles]
# Prepare the output tensor.
_, channels, height, width = image_tensor.shape
output_tensor = torch.zeros(
(height * scale, width * scale, channels), dtype=torch.uint8, device=torch.device("cpu")
)
image_tensor = image_tensor.to(device=spandrel_model.device, dtype=spandrel_model.dtype)
for tile, scaled_tile in tqdm(list(zip(tiles, scaled_tiles, strict=True)), desc="Upscaling Tiles"):
# Exit early if the invocation has been canceled.
if context.util.is_canceled():
raise CanceledException
# Extract the current tile from the input tensor.
input_tile = image_tensor[
:, :, tile.coords.top : tile.coords.bottom, tile.coords.left : tile.coords.right
].to(device=spandrel_model.device, dtype=spandrel_model.dtype)
# Run the model on the tile.
output_tile = spandrel_model.run(input_tile)
# Convert the output tile into the output tensor's format.
# (N, C, H, W) -> (C, H, W)
output_tile = output_tile.squeeze(0)
# (C, H, W) -> (H, W, C)
output_tile = output_tile.permute(1, 2, 0)
output_tile = output_tile.clamp(0, 1)
output_tile = (output_tile * 255).to(dtype=torch.uint8, device=torch.device("cpu"))
# Merge the output tile into the output tensor.
# We only keep half of the overlap on the top and left side of the tile. We do this in case there are
# edge artifacts. We don't bother with any 'blending' in the current implementation - for most upscalers
# it seems unnecessary, but we may find a need in the future.
top_overlap = scaled_tile.overlap.top // 2
left_overlap = scaled_tile.overlap.left // 2
output_tensor[
scaled_tile.coords.top + top_overlap : scaled_tile.coords.bottom,
scaled_tile.coords.left + left_overlap : scaled_tile.coords.right,
:,
] = output_tile[top_overlap:, left_overlap:, :]
# Convert the output tensor to a PIL image.
np_image = output_tensor.detach().numpy().astype(np.uint8)
pil_image = Image.fromarray(np_image)
image_dto = context.images.save(image=pil_image)
return ImageOutput.build(image_dto)

470
invokeai/app/services/workflow_records/default_workflows/ESRGAN Upscaling with Canny ControlNet.json
View File

@ -2,7 +2,7 @@
"name": "ESRGAN Upscaling with Canny ControlNet", "name": "ESRGAN Upscaling with Canny ControlNet",
"author": "InvokeAI", "author": "InvokeAI",
"description": "Sample workflow for using Upscaling with ControlNet with SD1.5", "description": "Sample workflow for using Upscaling with ControlNet with SD1.5",
"version": "2.0.0", "version": "2.1.0",
"contact": "invoke@invoke.ai", "contact": "invoke@invoke.ai",
"tags": "upscale, controlnet, default", "tags": "upscale, controlnet, default",
"notes": "", "notes": "",
@ -36,14 +36,13 @@
"version": "3.0.0", "version": "3.0.0",
"category": "default" "category": "default"
}, },
"id": "0e71a27e-a22b-4a9b-b20a-6d789abff2bc",
"nodes": [ "nodes": [
{ {
"id": "e8bf67fe-67de-4227-87eb-79e86afdfc74", "id": "63b6ab7e-5b05-4d1b-a3b1-42d8e53ce16b",
"type": "invocation", "type": "invocation",
"data": { "data": {
"id": "e8bf67fe-67de-4227-87eb-79e86afdfc74", "id": "63b6ab7e-5b05-4d1b-a3b1-42d8e53ce16b",
"version": "1.1.1", "version": "1.2.0",
"nodePack": "invokeai", "nodePack": "invokeai",
"label": "", "label": "",
"notes": "", "notes": "",
@ -57,6 +56,10 @@
"clip": { "clip": {
"name": "clip", "name": "clip",
"label": "" "label": ""
},
"mask": {
"name": "mask",
"label": ""
} }
}, },
"isOpen": true, "isOpen": true,
@ -65,122 +68,63 @@
}, },
"position": { "position": {
"x": 1250, "x": 1250,
"y": 1500 "y": 1200
} }
}, },
{ {
"id": "d8ace142-c05f-4f1d-8982-88dc7473958d", "id": "5ca498a4-c8c8-4580-a396-0c984317205d",
"type": "invocation", "type": "invocation",
"data": { "data": {
"id": "d8ace142-c05f-4f1d-8982-88dc7473958d", "id": "5ca498a4-c8c8-4580-a396-0c984317205d",
"version": "1.0.2", "version": "1.1.0",
"nodePack": "invokeai", "nodePack": "invokeai",
"label": "", "label": "",
"notes": "", "notes": "",
"type": "main_model_loader", "type": "i2l",
"inputs": { "inputs": {
"model": { "image": {
"name": "model", "name": "image",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "", "label": "",
"value": { "value": false
"key": "5cd43ca0-dd0a-418d-9f7e-35b2b9d5e106",
"hash": "blake3:6987f323017f597213cc3264250edf57056d21a40a0a85d83a1a33a7d44dc41a",
"name": "Deliberate_v5",
"base": "sd-1",
"type": "main"
}
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 700,
"y": 1375
}
},
{
"id": "771bdf6a-0813-4099-a5d8-921a138754d4",
"type": "invocation",
"data": {
"id": "771bdf6a-0813-4099-a5d8-921a138754d4",
"version": "1.0.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "image",
"inputs": {
"image": {
"name": "image",
"label": "Image To Upscale",
"value": {
"image_name": "d2e42ba6-d420-496b-82db-91c9b75956c1.png"
}
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 344.5593065887157,
"y": 1698.161491368619
}
},
{
"id": "f7564dd2-9539-47f2-ac13-190804461f4e",
"type": "invocation",
"data": {
"id": "f7564dd2-9539-47f2-ac13-190804461f4e",
"version": "1.3.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "esrgan",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"image": {
"name": "image",
"label": ""
},
"model_name": {
"name": "model_name",
"label": "Upscaler Model",
"value": "RealESRGAN_x2plus.pth"
}, },
"tile_size": { "tile_size": {
"name": "tile_size", "name": "tile_size",
"label": "", "label": "",
"value": 400 "value": 0
},
"fp32": {
"name": "fp32",
"label": "",
"value": false
} }
}, },
"isOpen": true, "isOpen": false,
"isIntermediate": true, "isIntermediate": true,
"useCache": true "useCache": true
}, },
"position": { "position": {
"x": 717.3863693661265, "x": 1650,
"y": 1721.9215053134815 "y": 1675
} }
}, },
{ {
"id": "1d887701-df21-4966-ae6e-a7d82307d7bd", "id": "3ed9b2ef-f4ec-40a7-94db-92e63b583ec0",
"type": "invocation", "type": "invocation",
"data": { "data": {
"id": "1d887701-df21-4966-ae6e-a7d82307d7bd", "id": "3ed9b2ef-f4ec-40a7-94db-92e63b583ec0",
"version": "1.3.2", "version": "1.3.0",
"nodePack": "invokeai", "nodePack": "invokeai",
"label": "", "label": "",
"notes": "", "notes": "",
"type": "canny_image_processor", "type": "l2i",
"inputs": { "inputs": {
"board": { "board": {
"name": "board", "name": "board",
@ -190,38 +134,37 @@
"name": "metadata", "name": "metadata",
"label": "" "label": ""
}, },
"image": { "latents": {
"name": "image", "name": "latents",
"label": "" "label": ""
}, },
"detect_resolution": { "vae": {
"name": "detect_resolution", "name": "vae",
"label": "", "label": ""
"value": 512
}, },
"image_resolution": { "tiled": {
"name": "image_resolution", "name": "tiled",
"label": "", "label": "",
"value": 512 "value": false
}, },
"low_threshold": { "tile_size": {
"name": "low_threshold", "name": "tile_size",
"label": "", "label": "",
"value": 100 "value": 0
}, },
"high_threshold": { "fp32": {
"name": "high_threshold", "name": "fp32",
"label": "", "label": "",
"value": 200 "value": false
} }
}, },
"isOpen": true, "isOpen": true,
"isIntermediate": true, "isIntermediate": false,
"useCache": true "useCache": true
}, },
"position": { "position": {
"x": 1200, "x": 2559.4751127537957,
"y": 1900 "y": 1246.6000376741406
} }
}, },
{ {
@ -229,7 +172,7 @@
"type": "invocation", "type": "invocation",
"data": { "data": {
"id": "ca1d020c-89a8-4958-880a-016d28775cfa", "id": "ca1d020c-89a8-4958-880a-016d28775cfa",
"version": "1.1.1", "version": "1.1.2",
"nodePack": "invokeai", "nodePack": "invokeai",
"label": "", "label": "",
"notes": "", "notes": "",
@ -285,6 +228,193 @@
"y": 1902.9649340196056 "y": 1902.9649340196056
} }
}, },
{
"id": "1d887701-df21-4966-ae6e-a7d82307d7bd",
"type": "invocation",
"data": {
"id": "1d887701-df21-4966-ae6e-a7d82307d7bd",
"version": "1.3.3",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "canny_image_processor",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"image": {
"name": "image",
"label": ""
},
"detect_resolution": {
"name": "detect_resolution",
"label": "",
"value": 512
},
"image_resolution": {
"name": "image_resolution",
"label": "",
"value": 512
},
"low_threshold": {
"name": "low_threshold",
"label": "",
"value": 100
},
"high_threshold": {
"name": "high_threshold",
"label": "",
"value": 200
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 1200,
"y": 1900
}
},
{
"id": "d8ace142-c05f-4f1d-8982-88dc7473958d",
"type": "invocation",
"data": {
"id": "d8ace142-c05f-4f1d-8982-88dc7473958d",
"version": "1.0.3",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "main_model_loader",
"inputs": {
"model": {
"name": "model",
"label": "",
"value": {
"key": "5cd43ca0-dd0a-418d-9f7e-35b2b9d5e106",
"hash": "blake3:6987f323017f597213cc3264250edf57056d21a40a0a85d83a1a33a7d44dc41a",
"name": "Deliberate_v5",
"base": "sd-1",
"type": "main"
}
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 700,
"y": 1375
}
},
{
"id": "e8bf67fe-67de-4227-87eb-79e86afdfc74",
"type": "invocation",
"data": {
"id": "e8bf67fe-67de-4227-87eb-79e86afdfc74",
"version": "1.2.0",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "compel",
"inputs": {
"prompt": {
"name": "prompt",
"label": "",
"value": ""
},
"clip": {
"name": "clip",
"label": ""
},
"mask": {
"name": "mask",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 1250,
"y": 1500
}
},
{
"id": "771bdf6a-0813-4099-a5d8-921a138754d4",
"type": "invocation",
"data": {
"id": "771bdf6a-0813-4099-a5d8-921a138754d4",
"version": "1.0.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "image",
"inputs": {
"image": {
"name": "image",
"label": "Image To Upscale"
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 344.5593065887157,
"y": 1698.161491368619
}
},
{
"id": "f7564dd2-9539-47f2-ac13-190804461f4e",
"type": "invocation",
"data": {
"id": "f7564dd2-9539-47f2-ac13-190804461f4e",
"version": "1.3.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "esrgan",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"image": {
"name": "image",
"label": ""
},
"model_name": {
"name": "model_name",
"label": "Upscaler Model",
"value": "RealESRGAN_x2plus.pth"
},
"tile_size": {
"name": "tile_size",
"label": "",
"value": 400
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 717.3863693661265,
"y": 1721.9215053134815
}
},
{ {
"id": "f50624ce-82bf-41d0-bdf7-8aab11a80d48", "id": "f50624ce-82bf-41d0-bdf7-8aab11a80d48",
"type": "invocation", "type": "invocation",
@ -413,122 +543,6 @@
"y": 1232.6219060454753 "y": 1232.6219060454753
} }
}, },
{
"id": "3ed9b2ef-f4ec-40a7-94db-92e63b583ec0",
"type": "invocation",
"data": {
"id": "3ed9b2ef-f4ec-40a7-94db-92e63b583ec0",
"version": "1.2.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "l2i",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"latents": {
"name": "latents",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"fp32": {
"name": "fp32",
"label": "",
"value": false
}
},
"isOpen": true,
"isIntermediate": false,
"useCache": true
},
"position": {
"x": 2559.4751127537957,
"y": 1246.6000376741406
}
},
{
"id": "5ca498a4-c8c8-4580-a396-0c984317205d",
"type": "invocation",
"data": {
"id": "5ca498a4-c8c8-4580-a396-0c984317205d",
"version": "1.0.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "i2l",
"inputs": {
"image": {
"name": "image",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"fp32": {
"name": "fp32",
"label": "",
"value": false
}
},
"isOpen": false,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 1650,
"y": 1675
}
},
{
"id": "63b6ab7e-5b05-4d1b-a3b1-42d8e53ce16b",
"type": "invocation",
"data": {
"id": "63b6ab7e-5b05-4d1b-a3b1-42d8e53ce16b",
"version": "1.1.1",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "compel",
"inputs": {
"prompt": {
"name": "prompt",
"label": "",
"value": ""
},
"clip": {
"name": "clip",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 1250,
"y": 1200
}
},
{ {
"id": "eb8f6f8a-c7b1-4914-806e-045ee2717a35", "id": "eb8f6f8a-c7b1-4914-806e-045ee2717a35",
"type": "invocation", "type": "invocation",

654
invokeai/app/services/workflow_records/default_workflows/Face Detailer with IP-Adapter & Canny (See Note in Details).json
View File

@ -2,7 +2,7 @@
"name": "Face Detailer with IP-Adapter & Canny (See Note in Details)", "name": "Face Detailer with IP-Adapter & Canny (See Note in Details)",
"author": "kosmoskatten", "author": "kosmoskatten",
"description": "A workflow to add detail to and improve faces. This workflow is most effective when used with a model that creates realistic outputs. ", "description": "A workflow to add detail to and improve faces. This workflow is most effective when used with a model that creates realistic outputs. ",
"version": "2.0.0", "version": "2.1.0",
"contact": "invoke@invoke.ai", "contact": "invoke@invoke.ai",
"tags": "face detailer, IP-Adapter, Canny", "tags": "face detailer, IP-Adapter, Canny",
"notes": "Set this image as the blur mask: https://i.imgur.com/Gxi61zP.png", "notes": "Set this image as the blur mask: https://i.imgur.com/Gxi61zP.png",
@ -37,16 +37,219 @@
} }
], ],
"meta": { "meta": {
"category": "default", "version": "3.0.0",
"version": "3.0.0" "category": "default"
}, },
"nodes": [ "nodes": [
{ {
"id": "44f2c190-eb03-460d-8d11-a94d13b33f19", "id": "c6359181-6479-40ec-bf3a-b7e8451683b8",
"type": "invocation", "type": "invocation",
"data": { "data": {
"id": "44f2c190-eb03-460d-8d11-a94d13b33f19", "id": "c6359181-6479-40ec-bf3a-b7e8451683b8",
"version": "1.1.1", "version": "1.0.3",
"label": "",
"notes": "",
"type": "main_model_loader",
"inputs": {
"model": {
"name": "model",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 2031.5518710051792,
"y": -492.1742944307074
}
},
{
"id": "8fe598c6-d447-44fa-a165-4975af77d080",
"type": "invocation",
"data": {
"id": "8fe598c6-d447-44fa-a165-4975af77d080",
"version": "1.3.3",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "canny_image_processor",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"image": {
"name": "image",
"label": ""
},
"detect_resolution": {
"name": "detect_resolution",
"label": "",
"value": 512
},
"image_resolution": {
"name": "image_resolution",
"label": "",
"value": 512
},
"low_threshold": {
"name": "low_threshold",
"label": "",
"value": 100
},
"high_threshold": {
"name": "high_threshold",
"label": "",
"value": 200
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 3519.4131037388597,
"y": 576.7946795840575
}
},
{
"id": "f60b6161-8f26-42f6-89ff-545e6011e501",
"type": "invocation",
"data": {
"id": "f60b6161-8f26-42f6-89ff-545e6011e501",
"version": "1.1.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "controlnet",
"inputs": {
"image": {
"name": "image",
"label": ""
},
"control_model": {
"name": "control_model",
"label": "Control Model (select canny)",
"value": {
"key": "5bdaacf7-a7a3-4fb8-b394-cc0ffbb8941d",
"hash": "blake3:260c7f8e10aefea9868cfc68d89970e91033bd37132b14b903e70ee05ebf530e",
"name": "sd-controlnet-canny",
"base": "sd-1",
"type": "controlnet"
}
},
"control_weight": {
"name": "control_weight",
"label": "",
"value": 0.5
},
"begin_step_percent": {
"name": "begin_step_percent",
"label": "",
"value": 0
},
"end_step_percent": {
"name": "end_step_percent",
"label": "",
"value": 0.5
},
"control_mode": {
"name": "control_mode",
"label": "",
"value": "balanced"
},
"resize_mode": {
"name": "resize_mode",
"label": "",
"value": "just_resize"
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 3950,
"y": 150
}
},
{
"id": "22b750db-b85e-486b-b278-ac983e329813",
"type": "invocation",
"data": {
"id": "22b750db-b85e-486b-b278-ac983e329813",
"version": "1.4.1",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "ip_adapter",
"inputs": {
"image": {
"name": "image",
"label": ""
},
"ip_adapter_model": {
"name": "ip_adapter_model",
"label": "IP-Adapter Model (select IP Adapter Face)",
"value": {
"key": "1cc210bb-4d0a-4312-b36c-b5d46c43768e",
"hash": "blake3:3d669dffa7471b357b4df088b99ffb6bf4d4383d5e0ef1de5ec1c89728a3d5a5",
"name": "ip_adapter_sd15",
"base": "sd-1",
"type": "ip_adapter"
}
},
"clip_vision_model": {
"name": "clip_vision_model",
"label": "",
"value": "ViT-H"
},
"weight": {
"name": "weight",
"label": "",
"value": 0.5
},
"method": {
"name": "method",
"label": "",
"value": "full"
},
"begin_step_percent": {
"name": "begin_step_percent",
"label": "",
"value": 0
},
"end_step_percent": {
"name": "end_step_percent",
"label": "",
"value": 0.8
},
"mask": {
"name": "mask",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 3575,
"y": -200
}
},
{
"id": "f4d15b64-c4a6-42a5-90fc-e4ed07a0ca65",
"type": "invocation",
"data": {
"id": "f4d15b64-c4a6-42a5-90fc-e4ed07a0ca65",
"version": "1.2.0",
"nodePack": "invokeai", "nodePack": "invokeai",
"label": "", "label": "",
"notes": "", "notes": "",
@ -60,6 +263,140 @@
"clip": { "clip": {
"name": "clip", "name": "clip",
"label": "" "label": ""
},
"mask": {
"name": "mask",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 2550,
"y": -525
}
},
{
"id": "2224ed72-2453-4252-bd89-3085240e0b6f",
"type": "invocation",
"data": {
"id": "2224ed72-2453-4252-bd89-3085240e0b6f",
"version": "1.3.0",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "l2i",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"latents": {
"name": "latents",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"tile_size": {
"name": "tile_size",
"label": "",
"value": 0
},
"fp32": {
"name": "fp32",
"label": "",
"value": true
}
},
"isOpen": true,
"isIntermediate": false,
"useCache": true
},
"position": {
"x": 4980.1395106966565,
"y": -255.9158921745602
}
},
{
"id": "de8b1a48-a2e4-42ca-90bb-66058bffd534",
"type": "invocation",
"data": {
"id": "de8b1a48-a2e4-42ca-90bb-66058bffd534",
"version": "1.1.0",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "i2l",
"inputs": {
"image": {
"name": "image",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"tile_size": {
"name": "tile_size",
"label": "",
"value": 0
},
"fp32": {
"name": "fp32",
"label": "",
"value": true
}
},
"isOpen": false,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 3100,
"y": -275
}
},
{
"id": "44f2c190-eb03-460d-8d11-a94d13b33f19",
"type": "invocation",
"data": {
"id": "44f2c190-eb03-460d-8d11-a94d13b33f19",
"version": "1.2.0",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "compel",
"inputs": {
"prompt": {
"name": "prompt",
"label": "",
"value": ""
},
"clip": {
"name": "clip",
"label": ""
},
"mask": {
"name": "mask",
"label": ""
} }
}, },
"isOpen": true, "isOpen": true,
@ -251,45 +588,6 @@
"y": 0 "y": 0
} }
}, },
{
"id": "de8b1a48-a2e4-42ca-90bb-66058bffd534",
"type": "invocation",
"data": {
"id": "de8b1a48-a2e4-42ca-90bb-66058bffd534",
"version": "1.0.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "i2l",
"inputs": {
"image": {
"name": "image",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"fp32": {
"name": "fp32",
"label": "",
"value": true
}
},
"isOpen": false,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 3100,
"y": -275
}
},
{ {
"id": "bd06261d-a74a-4d1f-8374-745ed6194bc2", "id": "bd06261d-a74a-4d1f-8374-745ed6194bc2",
"type": "invocation", "type": "invocation",
@ -418,53 +716,6 @@
"y": -175 "y": -175
} }
}, },
{
"id": "2224ed72-2453-4252-bd89-3085240e0b6f",
"type": "invocation",
"data": {
"id": "2224ed72-2453-4252-bd89-3085240e0b6f",
"version": "1.2.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "l2i",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"latents": {
"name": "latents",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"fp32": {
"name": "fp32",
"label": "",
"value": true
}
},
"isOpen": true,
"isIntermediate": false,
"useCache": true
},
"position": {
"x": 4980.1395106966565,
"y": -255.9158921745602
}
},
{ {
"id": "2974e5b3-3d41-4b6f-9953-cd21e8f3a323", "id": "2974e5b3-3d41-4b6f-9953-cd21e8f3a323",
"type": "invocation", "type": "invocation",
@ -692,201 +943,6 @@
"y": -275 "y": -275
} }
}, },
{
"id": "f4d15b64-c4a6-42a5-90fc-e4ed07a0ca65",
"type": "invocation",
"data": {
"id": "f4d15b64-c4a6-42a5-90fc-e4ed07a0ca65",
"version": "1.1.1",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "compel",
"inputs": {
"prompt": {
"name": "prompt",
"label": "",
"value": ""
},
"clip": {
"name": "clip",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 2550,
"y": -525
}
},
{
"id": "22b750db-b85e-486b-b278-ac983e329813",
"type": "invocation",
"data": {
"id": "22b750db-b85e-486b-b278-ac983e329813",
"version": "1.2.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "ip_adapter",
"inputs": {
"image": {
"name": "image",
"label": ""
},
"ip_adapter_model": {
"name": "ip_adapter_model",
"label": "IP-Adapter Model (select IP Adapter Face)",
"value": {
"key": "1cc210bb-4d0a-4312-b36c-b5d46c43768e",
"hash": "blake3:3d669dffa7471b357b4df088b99ffb6bf4d4383d5e0ef1de5ec1c89728a3d5a5",
"name": "ip_adapter_sd15",
"base": "sd-1",
"type": "ip_adapter"
}
},
"weight": {
"name": "weight",
"label": "",
"value": 0.5
},
"begin_step_percent": {
"name": "begin_step_percent",
"label": "",
"value": 0
},
"end_step_percent": {
"name": "end_step_percent",
"label": "",
"value": 0.8
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 3575,
"y": -200
}
},
{
"id": "f60b6161-8f26-42f6-89ff-545e6011e501",
"type": "invocation",
"data": {
"id": "f60b6161-8f26-42f6-89ff-545e6011e501",
"version": "1.1.1",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "controlnet",
"inputs": {
"image": {
"name": "image",
"label": ""
},
"control_model": {
"name": "control_model",
"label": "Control Model (select canny)",
"value": {
"key": "5bdaacf7-a7a3-4fb8-b394-cc0ffbb8941d",
"hash": "blake3:260c7f8e10aefea9868cfc68d89970e91033bd37132b14b903e70ee05ebf530e",
"name": "sd-controlnet-canny",
"base": "sd-1",
"type": "controlnet"
}
},
"control_weight": {
"name": "control_weight",
"label": "",
"value": 0.5
},
"begin_step_percent": {
"name": "begin_step_percent",
"label": "",
"value": 0
},
"end_step_percent": {
"name": "end_step_percent",
"label": "",
"value": 0.5
},
"control_mode": {
"name": "control_mode",
"label": "",
"value": "balanced"
},
"resize_mode": {
"name": "resize_mode",
"label": "",
"value": "just_resize"
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 3950,
"y": 150
}
},
{
"id": "8fe598c6-d447-44fa-a165-4975af77d080",
"type": "invocation",
"data": {
"id": "8fe598c6-d447-44fa-a165-4975af77d080",
"version": "1.3.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "canny_image_processor",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"image": {
"name": "image",
"label": ""
},
"detect_resolution": {
"name": "detect_resolution",
"label": "",
"value": 512
},
"image_resolution": {
"name": "image_resolution",
"label": "",
"value": 512
},
"low_threshold": {
"name": "low_threshold",
"label": "",
"value": 100
},
"high_threshold": {
"name": "high_threshold",
"label": "",
"value": 200
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 3519.4131037388597,
"y": 576.7946795840575
}
},
{ {
"id": "4bd4ae80-567f-4366-b8c6-3bb06f4fb46a", "id": "4bd4ae80-567f-4366-b8c6-3bb06f4fb46a",
"type": "invocation", "type": "invocation",
@ -1035,30 +1091,6 @@
"x": 2578.2364832140506, "x": 2578.2364832140506,
"y": 78.7948456497351 "y": 78.7948456497351
} }
},
{
"id": "c6359181-6479-40ec-bf3a-b7e8451683b8",
"type": "invocation",
"data": {
"id": "c6359181-6479-40ec-bf3a-b7e8451683b8",
"version": "1.0.2",
"label": "",
"notes": "",
"type": "main_model_loader",
"inputs": {
"model": {
"name": "model",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 2031.5518710051792,
"y": -492.1742944307074
}
} }
], ],
"edges": [ "edges": [

551
invokeai/app/services/workflow_records/default_workflows/Multi ControlNet (Canny & Depth).json
View File

@ -2,7 +2,7 @@
"name": "Multi ControlNet (Canny & Depth)", "name": "Multi ControlNet (Canny & Depth)",
"author": "InvokeAI", "author": "InvokeAI",
"description": "A sample workflow using canny & depth ControlNets to guide the generation process. ", "description": "A sample workflow using canny & depth ControlNets to guide the generation process. ",
"version": "2.0.0", "version": "2.1.0",
"contact": "invoke@invoke.ai", "contact": "invoke@invoke.ai",
"tags": "ControlNet, canny, depth", "tags": "ControlNet, canny, depth",
"notes": "", "notes": "",
@ -37,140 +37,104 @@
} }
], ],
"meta": { "meta": {
"category": "default", "version": "3.0.0",
"version": "3.0.0" "category": "default"
}, },
"nodes": [ "nodes": [
{ {
"id": "8e860e51-5045-456e-bf04-9a62a2a5c49e", "id": "9db25398-c869-4a63-8815-c6559341ef12",
"type": "invocation", "type": "invocation",
"data": { "data": {
"id": "8e860e51-5045-456e-bf04-9a62a2a5c49e", "id": "9db25398-c869-4a63-8815-c6559341ef12",
"version": "1.0.2", "version": "1.3.0",
"nodePack": "invokeai", "nodePack": "invokeai",
"label": "", "label": "",
"notes": "", "notes": "",
"type": "image", "type": "l2i",
"inputs": { "inputs": {
"image": { "board": {
"name": "image", "name": "board",
"label": "Depth Input Image"
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 3666.135718057363,
"y": 186.66887319822808
}
},
{
"id": "a33199c2-8340-401e-b8a2-42ffa875fc1c",
"type": "invocation",
"data": {
"id": "a33199c2-8340-401e-b8a2-42ffa875fc1c",
"version": "1.1.1",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "controlnet",
"inputs": {
"image": {
"name": "image",
"label": "" "label": ""
}, },
"control_model": { "metadata": {
"name": "control_model", "name": "metadata",
"label": "Control Model (select depth)", "label": ""
"value": {
"key": "87e8855c-671f-4c9e-bbbb-8ed47ccb4aac",
"hash": "blake3:2550bf22a53942dfa28ab2fed9d10d80851112531f44d977168992edf9d0534c",
"name": "control_v11f1p_sd15_depth",
"base": "sd-1",
"type": "controlnet"
}
}, },
"control_weight": { "latents": {
"name": "control_weight", "name": "latents",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "", "label": "",
"value": 1 "value": false
}, },
"begin_step_percent": { "tile_size": {
"name": "begin_step_percent", "name": "tile_size",
"label": "", "label": "",
"value": 0 "value": 0
}, },
"end_step_percent": { "fp32": {
"name": "end_step_percent", "name": "fp32",
"label": "", "label": "",
"value": 1 "value": false
},
"control_mode": {
"name": "control_mode",
"label": "",
"value": "balanced"
},
"resize_mode": {
"name": "resize_mode",
"label": "",
"value": "just_resize"
} }
}, },
"isOpen": true, "isOpen": true,
"isIntermediate": true, "isIntermediate": false,
"useCache": true "useCache": true
}, },
"position": { "position": {
"x": 4477.604342844504, "x": 5675,
"y": -49.39005411272677
}
},
{
"id": "273e3f96-49ea-4dc5-9d5b-9660390f14e1",
"type": "invocation",
"data": {
"id": "273e3f96-49ea-4dc5-9d5b-9660390f14e1",
"version": "1.1.1",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "compel",
"inputs": {
"prompt": {
"name": "prompt",
"label": "Negative Prompt",
"value": ""
},
"clip": {
"name": "clip",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 4075,
"y": -825 "y": -825
} }
}, },
{ {
"id": "54486974-835b-4d81-8f82-05f9f32ce9e9", "id": "c826ba5e-9676-4475-b260-07b85e88753c",
"type": "invocation", "type": "invocation",
"data": { "data": {
"id": "54486974-835b-4d81-8f82-05f9f32ce9e9", "id": "c826ba5e-9676-4475-b260-07b85e88753c",
"version": "1.0.2", "version": "1.3.3",
"nodePack": "invokeai", "nodePack": "invokeai",
"label": "", "label": "",
"notes": "", "notes": "",
"type": "main_model_loader", "type": "canny_image_processor",
"inputs": { "inputs": {
"model": { "board": {
"name": "model", "name": "board",
"label": "" "label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"image": {
"name": "image",
"label": ""
},
"detect_resolution": {
"name": "detect_resolution",
"label": "",
"value": 512
},
"image_resolution": {
"name": "image_resolution",
"label": "",
"value": 512
},
"low_threshold": {
"name": "low_threshold",
"label": "",
"value": 100
},
"high_threshold": {
"name": "high_threshold",
"label": "",
"value": 200
} }
}, },
"isOpen": true, "isOpen": true,
@ -178,29 +142,52 @@
"useCache": true "useCache": true
}, },
"position": { "position": {
"x": 3600, "x": 4095.757337055795,
"y": -1000 "y": -455.63440891935863
} }
}, },
{ {
"id": "7ce68934-3419-42d4-ac70-82cfc9397306", "id": "018b1214-c2af-43a7-9910-fb687c6726d7",
"type": "invocation", "type": "invocation",
"data": { "data": {
"id": "7ce68934-3419-42d4-ac70-82cfc9397306", "id": "018b1214-c2af-43a7-9910-fb687c6726d7",
"version": "1.1.1", "version": "1.2.4",
"nodePack": "invokeai", "nodePack": "invokeai",
"label": "", "label": "",
"notes": "", "notes": "",
"type": "compel", "type": "midas_depth_image_processor",
"inputs": { "inputs": {
"prompt": { "board": {
"name": "prompt", "name": "board",
"label": "Positive Prompt",
"value": ""
},
"clip": {
"name": "clip",
"label": "" "label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"image": {
"name": "image",
"label": ""
},
"a_mult": {
"name": "a_mult",
"label": "",
"value": 2
},
"bg_th": {
"name": "bg_th",
"label": "",
"value": 0.1
},
"detect_resolution": {
"name": "detect_resolution",
"label": "",
"value": 512
},
"image_resolution": {
"name": "image_resolution",
"label": "",
"value": 512
} }
}, },
"isOpen": true, "isOpen": true,
@ -208,8 +195,8 @@
"useCache": true "useCache": true
}, },
"position": { "position": {
"x": 4075, "x": 4082.783145980783,
"y": -1125 "y": 0.01629251229994111
} }
}, },
{ {
@ -217,7 +204,7 @@
"type": "invocation", "type": "invocation",
"data": { "data": {
"id": "d204d184-f209-4fae-a0a1-d152800844e1", "id": "d204d184-f209-4fae-a0a1-d152800844e1",
"version": "1.1.1", "version": "1.1.2",
"nodePack": "invokeai", "nodePack": "invokeai",
"label": "", "label": "",
"notes": "", "notes": "",
@ -273,6 +260,185 @@
"y": -618.4221638099414 "y": -618.4221638099414
} }
}, },
{
"id": "7ce68934-3419-42d4-ac70-82cfc9397306",
"type": "invocation",
"data": {
"id": "7ce68934-3419-42d4-ac70-82cfc9397306",
"version": "1.2.0",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "compel",
"inputs": {
"prompt": {
"name": "prompt",
"label": "Positive Prompt",
"value": ""
},
"clip": {
"name": "clip",
"label": ""
},
"mask": {
"name": "mask",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 4075,
"y": -1125
}
},
{
"id": "54486974-835b-4d81-8f82-05f9f32ce9e9",
"type": "invocation",
"data": {
"id": "54486974-835b-4d81-8f82-05f9f32ce9e9",
"version": "1.0.3",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "main_model_loader",
"inputs": {
"model": {
"name": "model",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 3600,
"y": -1000
}
},
{
"id": "273e3f96-49ea-4dc5-9d5b-9660390f14e1",
"type": "invocation",
"data": {
"id": "273e3f96-49ea-4dc5-9d5b-9660390f14e1",
"version": "1.2.0",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "compel",
"inputs": {
"prompt": {
"name": "prompt",
"label": "Negative Prompt",
"value": ""
},
"clip": {
"name": "clip",
"label": ""
},
"mask": {
"name": "mask",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 4075,
"y": -825
}
},
{
"id": "a33199c2-8340-401e-b8a2-42ffa875fc1c",
"type": "invocation",
"data": {
"id": "a33199c2-8340-401e-b8a2-42ffa875fc1c",
"version": "1.1.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "controlnet",
"inputs": {
"image": {
"name": "image",
"label": ""
},
"control_model": {
"name": "control_model",
"label": "Control Model (select depth)",
"value": {
"key": "87e8855c-671f-4c9e-bbbb-8ed47ccb4aac",
"hash": "blake3:2550bf22a53942dfa28ab2fed9d10d80851112531f44d977168992edf9d0534c",
"name": "control_v11f1p_sd15_depth",
"base": "sd-1",
"type": "controlnet"
}
},
"control_weight": {
"name": "control_weight",
"label": "",
"value": 1
},
"begin_step_percent": {
"name": "begin_step_percent",
"label": "",
"value": 0
},
"end_step_percent": {
"name": "end_step_percent",
"label": "",
"value": 1
},
"control_mode": {
"name": "control_mode",
"label": "",
"value": "balanced"
},
"resize_mode": {
"name": "resize_mode",
"label": "",
"value": "just_resize"
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 4477.604342844504,
"y": -49.39005411272677
}
},
{
"id": "8e860e51-5045-456e-bf04-9a62a2a5c49e",
"type": "invocation",
"data": {
"id": "8e860e51-5045-456e-bf04-9a62a2a5c49e",
"version": "1.0.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "image",
"inputs": {
"image": {
"name": "image",
"label": "Depth Input Image"
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 3666.135718057363,
"y": 186.66887319822808
}
},
{ {
"id": "c4b23e64-7986-40c4-9cad-46327b12e204", "id": "c4b23e64-7986-40c4-9cad-46327b12e204",
"type": "invocation", "type": "invocation",
@ -322,159 +488,6 @@
"y": -575 "y": -575
} }
}, },
{
"id": "018b1214-c2af-43a7-9910-fb687c6726d7",
"type": "invocation",
"data": {
"id": "018b1214-c2af-43a7-9910-fb687c6726d7",
"version": "1.2.3",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "midas_depth_image_processor",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"image": {
"name": "image",
"label": ""
},
"a_mult": {
"name": "a_mult",
"label": "",
"value": 2
},
"bg_th": {
"name": "bg_th",
"label": "",
"value": 0.1
},
"detect_resolution": {
"name": "detect_resolution",
"label": "",
"value": 512
},
"image_resolution": {
"name": "image_resolution",
"label": "",
"value": 512
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 4082.783145980783,
"y": 0.01629251229994111
}
},
{
"id": "c826ba5e-9676-4475-b260-07b85e88753c",
"type": "invocation",
"data": {
"id": "c826ba5e-9676-4475-b260-07b85e88753c",
"version": "1.3.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "canny_image_processor",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"image": {
"name": "image",
"label": ""
},
"detect_resolution": {
"name": "detect_resolution",
"label": "",
"value": 512
},
"image_resolution": {
"name": "image_resolution",
"label": "",
"value": 512
},
"low_threshold": {
"name": "low_threshold",
"label": "",
"value": 100
},
"high_threshold": {
"name": "high_threshold",
"label": "",
"value": 200
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 4095.757337055795,
"y": -455.63440891935863
}
},
{
"id": "9db25398-c869-4a63-8815-c6559341ef12",
"type": "invocation",
"data": {
"id": "9db25398-c869-4a63-8815-c6559341ef12",
"version": "1.2.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "l2i",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"latents": {
"name": "latents",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"fp32": {
"name": "fp32",
"label": "",
"value": false
}
},
"isOpen": true,
"isIntermediate": false,
"useCache": true
},
"position": {
"x": 5675,
"y": -825
}
},
{ {
"id": "ac481b7f-08bf-4a9d-9e0c-3a82ea5243ce", "id": "ac481b7f-08bf-4a9d-9e0c-3a82ea5243ce",
"type": "invocation", "type": "invocation",

2181
invokeai/app/services/workflow_records/default_workflows/MultiDiffusion SDXL (Beta).json Normal file
View File

File diff suppressed because it is too large Load Diff

229
invokeai/app/services/workflow_records/default_workflows/Prompt from File.json
View File

@ -2,7 +2,7 @@
"name": "Prompt from File", "name": "Prompt from File",
"author": "InvokeAI", "author": "InvokeAI",
"description": "Sample workflow using Prompt from File node", "description": "Sample workflow using Prompt from File node",
"version": "2.0.0", "version": "2.1.0",
"contact": "invoke@invoke.ai", "contact": "invoke@invoke.ai",
"tags": "text2image, prompt from file, default", "tags": "text2image, prompt from file, default",
"notes": "", "notes": "",
@ -37,16 +37,68 @@
} }
], ],
"meta": { "meta": {
"category": "default", "version": "3.0.0",
"version": "3.0.0" "category": "default"
}, },
"nodes": [ "nodes": [
{ {
"id": "c2eaf1ba-5708-4679-9e15-945b8b432692", "id": "491ec988-3c77-4c37-af8a-39a0c4e7a2a1",
"type": "invocation", "type": "invocation",
"data": { "data": {
"id": "c2eaf1ba-5708-4679-9e15-945b8b432692", "id": "491ec988-3c77-4c37-af8a-39a0c4e7a2a1",
"version": "1.1.1", "version": "1.3.0",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "l2i",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"latents": {
"name": "latents",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"tile_size": {
"name": "tile_size",
"label": "",
"value": 0
},
"fp32": {
"name": "fp32",
"label": "",
"value": false
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 2037.861329274915,
"y": -329.8393457509562
}
},
{
"id": "fc9d0e35-a6de-4a19-84e1-c72497c823f6",
"type": "invocation",
"data": {
"id": "fc9d0e35-a6de-4a19-84e1-c72497c823f6",
"version": "1.2.0",
"nodePack": "invokeai", "nodePack": "invokeai",
"label": "", "label": "",
"notes": "", "notes": "",
@ -60,6 +112,69 @@
"clip": { "clip": {
"name": "clip", "name": "clip",
"label": "" "label": ""
},
"mask": {
"name": "mask",
"label": ""
}
},
"isOpen": false,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 925,
"y": -275
}
},
{
"id": "d6353b7f-b447-4e17-8f2e-80a88c91d426",
"type": "invocation",
"data": {
"id": "d6353b7f-b447-4e17-8f2e-80a88c91d426",
"version": "1.0.3",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "main_model_loader",
"inputs": {
"model": {
"name": "model",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 0,
"y": -375
}
},
{
"id": "c2eaf1ba-5708-4679-9e15-945b8b432692",
"type": "invocation",
"data": {
"id": "c2eaf1ba-5708-4679-9e15-945b8b432692",
"version": "1.2.0",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "compel",
"inputs": {
"prompt": {
"name": "prompt",
"label": "",
"value": ""
},
"clip": {
"name": "clip",
"label": ""
},
"mask": {
"name": "mask",
"label": ""
} }
}, },
"isOpen": false, "isOpen": false,
@ -141,61 +256,6 @@
"y": -400 "y": -400
} }
}, },
{
"id": "d6353b7f-b447-4e17-8f2e-80a88c91d426",
"type": "invocation",
"data": {
"id": "d6353b7f-b447-4e17-8f2e-80a88c91d426",
"version": "1.0.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "main_model_loader",
"inputs": {
"model": {
"name": "model",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 0,
"y": -375
}
},
{
"id": "fc9d0e35-a6de-4a19-84e1-c72497c823f6",
"type": "invocation",
"data": {
"id": "fc9d0e35-a6de-4a19-84e1-c72497c823f6",
"version": "1.1.1",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "compel",
"inputs": {
"prompt": {
"name": "prompt",
"label": "",
"value": ""
},
"clip": {
"name": "clip",
"label": ""
}
},
"isOpen": false,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 925,
"y": -275
}
},
{ {
"id": "0eb5f3f5-1b91-49eb-9ef0-41d67c7eae77", "id": "0eb5f3f5-1b91-49eb-9ef0-41d67c7eae77",
"type": "invocation", "type": "invocation",
@ -268,53 +328,6 @@
"y": -50 "y": -50
} }
}, },
{
"id": "491ec988-3c77-4c37-af8a-39a0c4e7a2a1",
"type": "invocation",
"data": {
"id": "491ec988-3c77-4c37-af8a-39a0c4e7a2a1",
"version": "1.2.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "l2i",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"latents": {
"name": "latents",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"fp32": {
"name": "fp32",
"label": "",
"value": false
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 2037.861329274915,
"y": -329.8393457509562
}
},
{ {
"id": "2fb1577f-0a56-4f12-8711-8afcaaaf1d5e", "id": "2fb1577f-0a56-4f12-8711-8afcaaaf1d5e",
"type": "invocation", "type": "invocation",

225
invokeai/app/services/workflow_records/default_workflows/Text to Image - SD1.5.json
View File

@ -2,7 +2,7 @@
"name": "Text to Image - SD1.5", "name": "Text to Image - SD1.5",
"author": "InvokeAI", "author": "InvokeAI",
"description": "Sample text to image workflow for Stable Diffusion 1.5/2", "description": "Sample text to image workflow for Stable Diffusion 1.5/2",
"version": "2.0.0", "version": "2.1.0",
"contact": "invoke@invoke.ai", "contact": "invoke@invoke.ai",
"tags": "text2image, SD1.5, SD2, default", "tags": "text2image, SD1.5, SD2, default",
"notes": "", "notes": "",
@ -33,16 +33,127 @@
} }
], ],
"meta": { "meta": {
"category": "default", "version": "3.0.0",
"version": "3.0.0" "category": "default"
}, },
"nodes": [ "nodes": [
{
"id": "58c957f5-0d01-41fc-a803-b2bbf0413d4f",
"type": "invocation",
"data": {
"id": "58c957f5-0d01-41fc-a803-b2bbf0413d4f",
"version": "1.3.0",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "l2i",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"latents": {
"name": "latents",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"tile_size": {
"name": "tile_size",
"label": "",
"value": 0
},
"fp32": {
"name": "fp32",
"label": "",
"value": true
}
},
"isOpen": true,
"isIntermediate": false,
"useCache": true
},
"position": {
"x": 1800,
"y": 25
}
},
{
"id": "7d8bf987-284f-413a-b2fd-d825445a5d6c",
"type": "invocation",
"data": {
"id": "7d8bf987-284f-413a-b2fd-d825445a5d6c",
"version": "1.2.0",
"nodePack": "invokeai",
"label": "Positive Compel Prompt",
"notes": "",
"type": "compel",
"inputs": {
"prompt": {
"name": "prompt",
"label": "Positive Prompt",
"value": "Super cute tiger cub, national geographic award-winning photograph"
},
"clip": {
"name": "clip",
"label": ""
},
"mask": {
"name": "mask",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 1000,
"y": 25
}
},
{
"id": "c8d55139-f380-4695-b7f2-8b3d1e1e3db8",
"type": "invocation",
"data": {
"id": "c8d55139-f380-4695-b7f2-8b3d1e1e3db8",
"version": "1.0.3",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "main_model_loader",
"inputs": {
"model": {
"name": "model",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 600,
"y": 25
}
},
{ {
"id": "93dc02a4-d05b-48ed-b99c-c9b616af3402", "id": "93dc02a4-d05b-48ed-b99c-c9b616af3402",
"type": "invocation", "type": "invocation",
"data": { "data": {
"id": "93dc02a4-d05b-48ed-b99c-c9b616af3402", "id": "93dc02a4-d05b-48ed-b99c-c9b616af3402",
"version": "1.1.1", "version": "1.2.0",
"nodePack": "invokeai", "nodePack": "invokeai",
"label": "Negative Compel Prompt", "label": "Negative Compel Prompt",
"notes": "", "notes": "",
@ -56,6 +167,10 @@
"clip": { "clip": {
"name": "clip", "name": "clip",
"label": "" "label": ""
},
"mask": {
"name": "mask",
"label": ""
} }
}, },
"isOpen": true, "isOpen": true,
@ -108,61 +223,6 @@
"y": 325 "y": 325
} }
}, },
{
"id": "c8d55139-f380-4695-b7f2-8b3d1e1e3db8",
"type": "invocation",
"data": {
"id": "c8d55139-f380-4695-b7f2-8b3d1e1e3db8",
"version": "1.0.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "main_model_loader",
"inputs": {
"model": {
"name": "model",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 600,
"y": 25
}
},
{
"id": "7d8bf987-284f-413a-b2fd-d825445a5d6c",
"type": "invocation",
"data": {
"id": "7d8bf987-284f-413a-b2fd-d825445a5d6c",
"version": "1.1.1",
"nodePack": "invokeai",
"label": "Positive Compel Prompt",
"notes": "",
"type": "compel",
"inputs": {
"prompt": {
"name": "prompt",
"label": "Positive Prompt",
"value": "Super cute tiger cub, national geographic award-winning photograph"
},
"clip": {
"name": "clip",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 1000,
"y": 25
}
},
{ {
"id": "ea94bc37-d995-4a83-aa99-4af42479f2f2", "id": "ea94bc37-d995-4a83-aa99-4af42479f2f2",
"type": "invocation", "type": "invocation",
@ -280,53 +340,6 @@
"x": 1400, "x": 1400,
"y": 25 "y": 25
} }
},
{
"id": "58c957f5-0d01-41fc-a803-b2bbf0413d4f",
"type": "invocation",
"data": {
"id": "58c957f5-0d01-41fc-a803-b2bbf0413d4f",
"version": "1.2.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "l2i",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"latents": {
"name": "latents",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"fp32": {
"name": "fp32",
"label": "",
"value": true
}
},
"isOpen": true,
"isIntermediate": false,
"useCache": true
},
"position": {
"x": 1800,
"y": 25
}
} }
], ],
"edges": [ "edges": [

515
invokeai/app/services/workflow_records/default_workflows/Text to Image - SDXL.json
View File

@ -2,7 +2,7 @@
"name": "Text to Image - SDXL", "name": "Text to Image - SDXL",
"author": "InvokeAI", "author": "InvokeAI",
"description": "Sample text to image workflow for SDXL", "description": "Sample text to image workflow for SDXL",
"version": "2.0.0", "version": "2.1.0",
"contact": "invoke@invoke.ai", "contact": "invoke@invoke.ai",
"tags": "text2image, SDXL, default", "tags": "text2image, SDXL, default",
"notes": "", "notes": "",
@ -29,10 +29,271 @@
} }
], ],
"meta": { "meta": {
"category": "default", "version": "3.0.0",
"version": "3.0.0" "category": "default"
}, },
"nodes": [ "nodes": [
{
"id": "0093692f-9cf4-454d-a5b8-62f0e3eb3bb8",
"type": "invocation",
"data": {
"id": "0093692f-9cf4-454d-a5b8-62f0e3eb3bb8",
"version": "1.0.3",
"label": "",
"notes": "",
"type": "vae_loader",
"inputs": {
"vae_model": {
"name": "vae_model",
"label": "VAE (use the FP16 model)",
"value": {
"key": "f20f9e5c-1bce-4c46-a84d-34ebfa7df069",
"hash": "blake3:9705ab1c31fa96b308734214fb7571a958621c7a9247eed82b7d277145f8d9fa",
"name": "sdxl-vae-fp16-fix",
"base": "sdxl",
"type": "vae"
}
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 375,
"y": -225
}
},
{
"id": "63e91020-83b2-4f35-b174-ad9692aabb48",
"type": "invocation",
"data": {
"id": "63e91020-83b2-4f35-b174-ad9692aabb48",
"version": "1.3.0",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "l2i",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"latents": {
"name": "latents",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"tile_size": {
"name": "tile_size",
"label": "",
"value": 0
},
"fp32": {
"name": "fp32",
"label": "",
"value": false
}
},
"isOpen": true,
"isIntermediate": false,
"useCache": false
},
"position": {
"x": 1475,
"y": -500
}
},
{
"id": "faf965a4-7530-427b-b1f3-4ba6505c2a08",
"type": "invocation",
"data": {
"id": "faf965a4-7530-427b-b1f3-4ba6505c2a08",
"version": "1.2.0",
"nodePack": "invokeai",
"label": "SDXL Positive Compel Prompt",
"notes": "",
"type": "sdxl_compel_prompt",
"inputs": {
"prompt": {
"name": "prompt",
"label": "Positive Prompt",
"value": ""
},
"style": {
"name": "style",
"label": "Positive Style",
"value": ""
},
"original_width": {
"name": "original_width",
"label": "",
"value": 1024
},
"original_height": {
"name": "original_height",
"label": "",
"value": 1024
},
"crop_top": {
"name": "crop_top",
"label": "",
"value": 0
},
"crop_left": {
"name": "crop_left",
"label": "",
"value": 0
},
"target_width": {
"name": "target_width",
"label": "",
"value": 1024
},
"target_height": {
"name": "target_height",
"label": "",
"value": 1024
},
"clip": {
"name": "clip",
"label": ""
},
"clip2": {
"name": "clip2",
"label": ""
},
"mask": {
"name": "mask",
"label": ""
}
},
"isOpen": false,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 750,
"y": -175
}
},
{
"id": "30d3289c-773c-4152-a9d2-bd8a99c8fd22",
"type": "invocation",
"data": {
"id": "30d3289c-773c-4152-a9d2-bd8a99c8fd22",
"version": "1.0.3",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "sdxl_model_loader",
"inputs": {
"model": {
"name": "model",
"label": "",
"value": {
"key": "4a63b226-e8ff-4da4-854e-0b9f04b562ba",
"hash": "blake3:d279309ea6e5ee6e8fd52504275865cc280dac71cbf528c5b07c98b888bddaba",
"name": "dreamshaper-xl-v2-turbo",
"base": "sdxl",
"type": "main"
}
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 375,
"y": -500
}
},
{
"id": "3193ad09-a7c2-4bf4-a3a9-1c61cc33a204",
"type": "invocation",
"data": {
"id": "3193ad09-a7c2-4bf4-a3a9-1c61cc33a204",
"version": "1.2.0",
"nodePack": "invokeai",
"label": "SDXL Negative Compel Prompt",
"notes": "",
"type": "sdxl_compel_prompt",
"inputs": {
"prompt": {
"name": "prompt",
"label": "Negative Prompt",
"value": ""
},
"style": {
"name": "style",
"label": "Negative Style",
"value": ""
},
"original_width": {
"name": "original_width",
"label": "",
"value": 1024
},
"original_height": {
"name": "original_height",
"label": "",
"value": 1024
},
"crop_top": {
"name": "crop_top",
"label": "",
"value": 0
},
"crop_left": {
"name": "crop_left",
"label": "",
"value": 0
},
"target_width": {
"name": "target_width",
"label": "",
"value": 1024
},
"target_height": {
"name": "target_height",
"label": "",
"value": 1024
},
"clip": {
"name": "clip",
"label": ""
},
"clip2": {
"name": "clip2",
"label": ""
},
"mask": {
"name": "mask",
"label": ""
}
},
"isOpen": false,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 750,
"y": 200
}
},
{ {
"id": "3774ec24-a69e-4254-864c-097d07a6256f", "id": "3774ec24-a69e-4254-864c-097d07a6256f",
"type": "invocation", "type": "invocation",
@ -88,75 +349,6 @@
"y": -125 "y": -125
} }
}, },
{
"id": "3193ad09-a7c2-4bf4-a3a9-1c61cc33a204",
"type": "invocation",
"data": {
"id": "3193ad09-a7c2-4bf4-a3a9-1c61cc33a204",
"version": "1.1.1",
"nodePack": "invokeai",
"label": "SDXL Negative Compel Prompt",
"notes": "",
"type": "sdxl_compel_prompt",
"inputs": {
"prompt": {
"name": "prompt",
"label": "Negative Prompt",
"value": ""
},
"style": {
"name": "style",
"label": "Negative Style",
"value": ""
},
"original_width": {
"name": "original_width",
"label": "",
"value": 1024
},
"original_height": {
"name": "original_height",
"label": "",
"value": 1024
},
"crop_top": {
"name": "crop_top",
"label": "",
"value": 0
},
"crop_left": {
"name": "crop_left",
"label": "",
"value": 0
},
"target_width": {
"name": "target_width",
"label": "",
"value": 1024
},
"target_height": {
"name": "target_height",
"label": "",
"value": 1024
},
"clip": {
"name": "clip",
"label": ""
},
"clip2": {
"name": "clip2",
"label": ""
}
},
"isOpen": false,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 750,
"y": 200
}
},
{ {
"id": "55705012-79b9-4aac-9f26-c0b10309785b", "id": "55705012-79b9-4aac-9f26-c0b10309785b",
"type": "invocation", "type": "invocation",
@ -229,154 +421,6 @@
"y": -50 "y": -50
} }
}, },
{
"id": "30d3289c-773c-4152-a9d2-bd8a99c8fd22",
"type": "invocation",
"data": {
"id": "30d3289c-773c-4152-a9d2-bd8a99c8fd22",
"version": "1.0.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "sdxl_model_loader",
"inputs": {
"model": {
"name": "model",
"label": "",
"value": {
"key": "4a63b226-e8ff-4da4-854e-0b9f04b562ba",
"hash": "blake3:d279309ea6e5ee6e8fd52504275865cc280dac71cbf528c5b07c98b888bddaba",
"name": "dreamshaper-xl-v2-turbo",
"base": "sdxl",
"type": "main"
}
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 375,
"y": -500
}
},
{
"id": "faf965a4-7530-427b-b1f3-4ba6505c2a08",
"type": "invocation",
"data": {
"id": "faf965a4-7530-427b-b1f3-4ba6505c2a08",
"version": "1.1.1",
"nodePack": "invokeai",
"label": "SDXL Positive Compel Prompt",
"notes": "",
"type": "sdxl_compel_prompt",
"inputs": {
"prompt": {
"name": "prompt",
"label": "Positive Prompt",
"value": ""
},
"style": {
"name": "style",
"label": "Positive Style",
"value": ""
},
"original_width": {
"name": "original_width",
"label": "",
"value": 1024
},
"original_height": {
"name": "original_height",
"label": "",
"value": 1024
},
"crop_top": {
"name": "crop_top",
"label": "",
"value": 0
},
"crop_left": {
"name": "crop_left",
"label": "",
"value": 0
},
"target_width": {
"name": "target_width",
"label": "",
"value": 1024
},
"target_height": {
"name": "target_height",
"label": "",
"value": 1024
},
"clip": {
"name": "clip",
"label": ""
},
"clip2": {
"name": "clip2",
"label": ""
}
},
"isOpen": false,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 750,
"y": -175
}
},
{
"id": "63e91020-83b2-4f35-b174-ad9692aabb48",
"type": "invocation",
"data": {
"id": "63e91020-83b2-4f35-b174-ad9692aabb48",
"version": "1.2.2",
"nodePack": "invokeai",
"label": "",
"notes": "",
"type": "l2i",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"latents": {
"name": "latents",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"fp32": {
"name": "fp32",
"label": "",
"value": false
}
},
"isOpen": true,
"isIntermediate": false,
"useCache": false
},
"position": {
"x": 1475,
"y": -500
}
},
{ {
"id": "50a36525-3c0a-4cc5-977c-e4bfc3fd6dfb", "id": "50a36525-3c0a-4cc5-977c-e4bfc3fd6dfb",
"type": "invocation", "type": "invocation",
@ -464,37 +508,6 @@
"y": -500 "y": -500
} }
}, },
{
"id": "0093692f-9cf4-454d-a5b8-62f0e3eb3bb8",
"type": "invocation",
"data": {
"id": "0093692f-9cf4-454d-a5b8-62f0e3eb3bb8",
"version": "1.0.2",
"label": "",
"notes": "",
"type": "vae_loader",
"inputs": {
"vae_model": {
"name": "vae_model",
"label": "VAE (use the FP16 model)",
"value": {
"key": "f20f9e5c-1bce-4c46-a84d-34ebfa7df069",
"hash": "blake3:9705ab1c31fa96b308734214fb7571a958621c7a9247eed82b7d277145f8d9fa",
"name": "sdxl-vae-fp16-fix",
"base": "sdxl",
"type": "vae"
}
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 375,
"y": -225
}
},
{ {
"id": "ade2c0d3-0384-4157-b39b-29ce429cfa15", "id": "ade2c0d3-0384-4157-b39b-29ce429cfa15",
"type": "invocation", "type": "invocation",

185
invokeai/app/services/workflow_records/default_workflows/Text to Image with LoRA.json
View File

@ -2,7 +2,7 @@
"name": "Text to Image with LoRA", "name": "Text to Image with LoRA",
"author": "InvokeAI", "author": "InvokeAI",
"description": "Simple text to image workflow with a LoRA", "description": "Simple text to image workflow with a LoRA",
"version": "2.0.0", "version": "2.1.0",
"contact": "invoke@invoke.ai", "contact": "invoke@invoke.ai",
"tags": "text to image, lora, default", "tags": "text to image, lora, default",
"notes": "", "notes": "",
@ -37,28 +37,83 @@
} }
], ],
"meta": { "meta": {
"category": "default", "version": "3.0.0",
"version": "3.0.0" "category": "default"
}, },
"nodes": [ "nodes": [
{ {
"id": "85b77bb2-c67a-416a-b3e8-291abe746c44", "id": "a9683c0a-6b1f-4a5e-8187-c57e764b3400",
"type": "invocation", "type": "invocation",
"data": { "data": {
"id": "85b77bb2-c67a-416a-b3e8-291abe746c44", "id": "a9683c0a-6b1f-4a5e-8187-c57e764b3400",
"version": "1.1.1", "version": "1.3.0",
"label": "",
"notes": "",
"type": "l2i",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"latents": {
"name": "latents",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"tile_size": {
"name": "tile_size",
"label": "",
"value": 0
},
"fp32": {
"name": "fp32",
"label": "",
"value": false
}
},
"isOpen": true,
"isIntermediate": false,
"useCache": true
},
"position": {
"x": 4450,
"y": -550
}
},
{
"id": "c3fa6872-2599-4a82-a596-b3446a66cf8b",
"type": "invocation",
"data": {
"id": "c3fa6872-2599-4a82-a596-b3446a66cf8b",
"version": "1.2.0",
"label": "", "label": "",
"notes": "", "notes": "",
"type": "compel", "type": "compel",
"inputs": { "inputs": {
"prompt": { "prompt": {
"name": "prompt", "name": "prompt",
"label": "Negative Prompt", "label": "Positive Prompt",
"value": "" "value": "super cute tiger cub"
}, },
"clip": { "clip": {
"name": "clip", "name": "clip",
"label": "" "label": ""
},
"mask": {
"name": "mask",
"label": ""
} }
}, },
"isOpen": true, "isOpen": true,
@ -67,31 +122,7 @@
}, },
"position": { "position": {
"x": 3425, "x": 3425,
"y": -300 "y": -575
}
},
{
"id": "24e9d7ed-4836-4ec4-8f9e-e747721f9818",
"type": "invocation",
"data": {
"id": "24e9d7ed-4836-4ec4-8f9e-e747721f9818",
"version": "1.0.2",
"label": "",
"notes": "",
"type": "main_model_loader",
"inputs": {
"model": {
"name": "model",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 2500,
"y": -600
} }
}, },
{ {
@ -99,7 +130,7 @@
"type": "invocation", "type": "invocation",
"data": { "data": {
"id": "c41e705b-f2e3-4d1a-83c4-e34bb9344966", "id": "c41e705b-f2e3-4d1a-83c4-e34bb9344966",
"version": "1.0.2", "version": "1.0.3",
"label": "", "label": "",
"notes": "", "notes": "",
"type": "lora_loader", "type": "lora_loader",
@ -132,23 +163,51 @@
} }
}, },
{ {
"id": "c3fa6872-2599-4a82-a596-b3446a66cf8b", "id": "24e9d7ed-4836-4ec4-8f9e-e747721f9818",
"type": "invocation", "type": "invocation",
"data": { "data": {
"id": "c3fa6872-2599-4a82-a596-b3446a66cf8b", "id": "24e9d7ed-4836-4ec4-8f9e-e747721f9818",
"version": "1.1.1", "version": "1.0.3",
"label": "",
"notes": "",
"type": "main_model_loader",
"inputs": {
"model": {
"name": "model",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": 2500,
"y": -600
}
},
{
"id": "85b77bb2-c67a-416a-b3e8-291abe746c44",
"type": "invocation",
"data": {
"id": "85b77bb2-c67a-416a-b3e8-291abe746c44",
"version": "1.2.0",
"label": "", "label": "",
"notes": "", "notes": "",
"type": "compel", "type": "compel",
"inputs": { "inputs": {
"prompt": { "prompt": {
"name": "prompt", "name": "prompt",
"label": "Positive Prompt", "label": "Negative Prompt",
"value": "super cute tiger cub" "value": ""
}, },
"clip": { "clip": {
"name": "clip", "name": "clip",
"label": "" "label": ""
},
"mask": {
"name": "mask",
"label": ""
} }
}, },
"isOpen": true, "isOpen": true,
@ -157,7 +216,7 @@
}, },
"position": { "position": {
"x": 3425, "x": 3425,
"y": -575 "y": -300
} }
}, },
{ {
@ -315,52 +374,6 @@
"x": 3425, "x": 3425,
"y": 0 "y": 0
} }
},
{
"id": "a9683c0a-6b1f-4a5e-8187-c57e764b3400",
"type": "invocation",
"data": {
"id": "a9683c0a-6b1f-4a5e-8187-c57e764b3400",
"version": "1.2.2",
"label": "",
"notes": "",
"type": "l2i",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"latents": {
"name": "latents",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"fp32": {
"name": "fp32",
"label": "",
"value": false
}
},
"isOpen": true,
"isIntermediate": false,
"useCache": true
},
"position": {
"x": 4450,
"y": -550
}
} }
], ],
"edges": [ "edges": [

590
invokeai/app/services/workflow_records/default_workflows/Tiled Upscaling (Beta).json
View File

@ -2,7 +2,7 @@
"name": "Tiled Upscaling (Beta)", "name": "Tiled Upscaling (Beta)",
"author": "Invoke", "author": "Invoke",
"description": "A workflow to upscale an input image with tiled upscaling. ", "description": "A workflow to upscale an input image with tiled upscaling. ",
"version": "2.0.0", "version": "2.1.0",
"contact": "invoke@invoke.ai", "contact": "invoke@invoke.ai",
"tags": "tiled, upscaling, sd1.5", "tags": "tiled, upscaling, sd1.5",
"notes": "", "notes": "",
@ -41,10 +41,318 @@
} }
], ],
"meta": { "meta": {
"category": "default", "version": "3.0.0",
"version": "3.0.0" "category": "default"
}, },
"nodes": [ "nodes": [
{
"id": "2ff466b8-5e2a-4d8f-923a-a3884c7ecbc5",
"type": "invocation",
"data": {
"id": "2ff466b8-5e2a-4d8f-923a-a3884c7ecbc5",
"version": "1.0.3",
"label": "",
"notes": "",
"type": "main_model_loader",
"inputs": {
"model": {
"name": "model",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": -4514.466823162653,
"y": -1235.7908800002283
}
},
{
"id": "287f134f-da8d-41d1-884e-5940e8f7b816",
"type": "invocation",
"data": {
"id": "287f134f-da8d-41d1-884e-5940e8f7b816",
"version": "1.4.1",
"label": "",
"notes": "",
"type": "ip_adapter",
"inputs": {
"image": {
"name": "image",
"label": ""
},
"ip_adapter_model": {
"name": "ip_adapter_model",
"label": "IP-Adapter Model (select ip_adapter_sd15)",
"value": {
"key": "1cc210bb-4d0a-4312-b36c-b5d46c43768e",
"hash": "blake3:3d669dffa7471b357b4df088b99ffb6bf4d4383d5e0ef1de5ec1c89728a3d5a5",
"name": "ip_adapter_sd15",
"base": "sd-1",
"type": "ip_adapter"
}
},
"clip_vision_model": {
"name": "clip_vision_model",
"label": "",
"value": "ViT-H"
},
"weight": {
"name": "weight",
"label": "",
"value": 0.2
},
"method": {
"name": "method",
"label": "",
"value": "full"
},
"begin_step_percent": {
"name": "begin_step_percent",
"label": "",
"value": 0
},
"end_step_percent": {
"name": "end_step_percent",
"label": "",
"value": 1
},
"mask": {
"name": "mask",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": -2855.8555540799207,
"y": -183.58854843775742
}
},
{
"id": "b76fe66f-7884-43ad-b72c-fadc81d7a73c",
"type": "invocation",
"data": {
"id": "b76fe66f-7884-43ad-b72c-fadc81d7a73c",
"version": "1.3.0",
"label": "",
"notes": "",
"type": "l2i",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"latents": {
"name": "latents",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"tile_size": {
"name": "tile_size",
"label": "",
"value": 0
},
"fp32": {
"name": "fp32",
"label": "",
"value": false
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": -1999.770193862987,
"y": -1075
}
},
{
"id": "d334f2da-016a-4524-9911-bdab85546888",
"type": "invocation",
"data": {
"id": "d334f2da-016a-4524-9911-bdab85546888",
"version": "1.1.2",
"label": "",
"notes": "",
"type": "controlnet",
"inputs": {
"image": {
"name": "image",
"label": ""
},
"control_model": {
"name": "control_model",
"label": "Control Model (select contro_v11f1e_sd15_tile)",
"value": {
"key": "773843c8-db1f-4502-8f65-59782efa7960",
"hash": "blake3:f0812e13758f91baf4e54b7dbb707b70642937d3b2098cd2b94cc36d3eba308e",
"name": "control_v11f1e_sd15_tile",
"base": "sd-1",
"type": "controlnet"
}
},
"control_weight": {
"name": "control_weight",
"label": "",
"value": 1
},
"begin_step_percent": {
"name": "begin_step_percent",
"label": "",
"value": 0
},
"end_step_percent": {
"name": "end_step_percent",
"label": "Structural Control",
"value": 1
},
"control_mode": {
"name": "control_mode",
"label": "",
"value": "more_control"
},
"resize_mode": {
"name": "resize_mode",
"label": "",
"value": "just_resize"
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": -2481.9569385477016,
"y": -181.06590482739782
}
},
{
"id": "338b883c-3728-4f18-b3a6-6e7190c2f850",
"type": "invocation",
"data": {
"id": "338b883c-3728-4f18-b3a6-6e7190c2f850",
"version": "1.1.0",
"label": "",
"notes": "",
"type": "i2l",
"inputs": {
"image": {
"name": "image",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"tile_size": {
"name": "tile_size",
"label": "",
"value": 0
},
"fp32": {
"name": "fp32",
"label": "",
"value": false
}
},
"isOpen": false,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": -2908.4791167517287,
"y": -408.87504820159086
}
},
{
"id": "947c3f88-0305-4695-8355-df4abac64b1c",
"type": "invocation",
"data": {
"id": "947c3f88-0305-4695-8355-df4abac64b1c",
"version": "1.2.0",
"label": "",
"notes": "",
"type": "compel",
"inputs": {
"prompt": {
"name": "prompt",
"label": "",
"value": ""
},
"clip": {
"name": "clip",
"label": ""
},
"mask": {
"name": "mask",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": -4014.4136788915944,
"y": -968.5677253775948
}
},
{
"id": "9b2d8c58-ce8f-4162-a5a1-48de854040d6",
"type": "invocation",
"data": {
"id": "9b2d8c58-ce8f-4162-a5a1-48de854040d6",
"version": "1.2.0",
"label": "",
"notes": "",
"type": "compel",
"inputs": {
"prompt": {
"name": "prompt",
"label": "Positive Prompt",
"value": ""
},
"clip": {
"name": "clip",
"label": ""
},
"mask": {
"name": "mask",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": -4014.4136788915944,
"y": -1243.5677253775948
}
},
{ {
"id": "b875cae6-d8a3-4fdc-b969-4d53cbd03f9a", "id": "b875cae6-d8a3-4fdc-b969-4d53cbd03f9a",
"type": "invocation", "type": "invocation",
@ -181,64 +489,6 @@
"y": 3.422855503409039 "y": 3.422855503409039
} }
}, },
{
"id": "9b2d8c58-ce8f-4162-a5a1-48de854040d6",
"type": "invocation",
"data": {
"id": "9b2d8c58-ce8f-4162-a5a1-48de854040d6",
"version": "1.1.1",
"label": "",
"notes": "",
"type": "compel",
"inputs": {
"prompt": {
"name": "prompt",
"label": "Positive Prompt",
"value": ""
},
"clip": {
"name": "clip",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": -4014.4136788915944,
"y": -1243.5677253775948
}
},
{
"id": "947c3f88-0305-4695-8355-df4abac64b1c",
"type": "invocation",
"data": {
"id": "947c3f88-0305-4695-8355-df4abac64b1c",
"version": "1.1.1",
"label": "",
"notes": "",
"type": "compel",
"inputs": {
"prompt": {
"name": "prompt",
"label": "",
"value": ""
},
"clip": {
"name": "clip",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": -4014.4136788915944,
"y": -968.5677253775948
}
},
{ {
"id": "b3513fed-ed42-408d-b382-128fdb0de523", "id": "b3513fed-ed42-408d-b382-128fdb0de523",
"type": "invocation", "type": "invocation",
@ -379,104 +629,6 @@
"y": -29.08699277598673 "y": -29.08699277598673
} }
}, },
{
"id": "338b883c-3728-4f18-b3a6-6e7190c2f850",
"type": "invocation",
"data": {
"id": "338b883c-3728-4f18-b3a6-6e7190c2f850",
"version": "1.0.2",
"label": "",
"notes": "",
"type": "i2l",
"inputs": {
"image": {
"name": "image",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"fp32": {
"name": "fp32",
"label": "",
"value": false
}
},
"isOpen": false,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": -2908.4791167517287,
"y": -408.87504820159086
}
},
{
"id": "d334f2da-016a-4524-9911-bdab85546888",
"type": "invocation",
"data": {
"id": "d334f2da-016a-4524-9911-bdab85546888",
"version": "1.1.1",
"label": "",
"notes": "",
"type": "controlnet",
"inputs": {
"image": {
"name": "image",
"label": ""
},
"control_model": {
"name": "control_model",
"label": "Control Model (select contro_v11f1e_sd15_tile)",
"value": {
"key": "773843c8-db1f-4502-8f65-59782efa7960",
"hash": "blake3:f0812e13758f91baf4e54b7dbb707b70642937d3b2098cd2b94cc36d3eba308e",
"name": "control_v11f1e_sd15_tile",
"base": "sd-1",
"type": "controlnet"
}
},
"control_weight": {
"name": "control_weight",
"label": "",
"value": 1
},
"begin_step_percent": {
"name": "begin_step_percent",
"label": "",
"value": 0
},
"end_step_percent": {
"name": "end_step_percent",
"label": "Structural Control",
"value": 1
},
"control_mode": {
"name": "control_mode",
"label": "",
"value": "more_control"
},
"resize_mode": {
"name": "resize_mode",
"label": "",
"value": "just_resize"
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": -2481.9569385477016,
"y": -181.06590482739782
}
},
{ {
"id": "1011539e-85de-4e02-a003-0b22358491b8", "id": "1011539e-85de-4e02-a003-0b22358491b8",
"type": "invocation", "type": "invocation",
@ -563,52 +715,6 @@
"y": -1006.415909408244 "y": -1006.415909408244
} }
}, },
{
"id": "b76fe66f-7884-43ad-b72c-fadc81d7a73c",
"type": "invocation",
"data": {
"id": "b76fe66f-7884-43ad-b72c-fadc81d7a73c",
"version": "1.2.2",
"label": "",
"notes": "",
"type": "l2i",
"inputs": {
"board": {
"name": "board",
"label": ""
},
"metadata": {
"name": "metadata",
"label": ""
},
"latents": {
"name": "latents",
"label": ""
},
"vae": {
"name": "vae",
"label": ""
},
"tiled": {
"name": "tiled",
"label": "",
"value": false
},
"fp32": {
"name": "fp32",
"label": "",
"value": false
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": -1999.770193862987,
"y": -1075
}
},
{ {
"id": "ab6f5dda-4b60-4ddf-99f2-f61fb5937527", "id": "ab6f5dda-4b60-4ddf-99f2-f61fb5937527",
"type": "invocation", "type": "invocation",
@ -779,56 +885,6 @@
"y": -78.2819050861178 "y": -78.2819050861178
} }
}, },
{
"id": "287f134f-da8d-41d1-884e-5940e8f7b816",
"type": "invocation",
"data": {
"id": "287f134f-da8d-41d1-884e-5940e8f7b816",
"version": "1.2.2",
"label": "",
"notes": "",
"type": "ip_adapter",
"inputs": {
"image": {
"name": "image",
"label": ""
},
"ip_adapter_model": {
"name": "ip_adapter_model",
"label": "IP-Adapter Model (select ip_adapter_sd15)",
"value": {
"key": "1cc210bb-4d0a-4312-b36c-b5d46c43768e",
"hash": "blake3:3d669dffa7471b357b4df088b99ffb6bf4d4383d5e0ef1de5ec1c89728a3d5a5",
"name": "ip_adapter_sd15",
"base": "sd-1",
"type": "ip_adapter"
}
},
"weight": {
"name": "weight",
"label": "",
"value": 0.2
},
"begin_step_percent": {
"name": "begin_step_percent",
"label": "",
"value": 0
},
"end_step_percent": {
"name": "end_step_percent",
"label": "",
"value": 1
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": -2855.8555540799207,
"y": -183.58854843775742
}
},
{ {
"id": "1f86c8bf-06f9-4e28-abee-02f46f445ac4", "id": "1f86c8bf-06f9-4e28-abee-02f46f445ac4",
"type": "invocation", "type": "invocation",
@ -899,30 +955,6 @@
"y": -41.810810454906914 "y": -41.810810454906914
} }
}, },
{
"id": "2ff466b8-5e2a-4d8f-923a-a3884c7ecbc5",
"type": "invocation",
"data": {
"id": "2ff466b8-5e2a-4d8f-923a-a3884c7ecbc5",
"version": "1.0.2",
"label": "",
"notes": "",
"type": "main_model_loader",
"inputs": {
"model": {
"name": "model",
"label": ""
}
},
"isOpen": true,
"isIntermediate": true,
"useCache": true
},
"position": {
"x": -4514.466823162653,
"y": -1235.7908800002283
}
},
{ {
"id": "f5d9bf3b-2646-4b17-9894-20fd2b4218ea", "id": "f5d9bf3b-2646-4b17-9894-20fd2b4218ea",
"type": "invocation", "type": "invocation",

2
invokeai/backend/image_util/lineart_anime.py
View File

@ -98,7 +98,7 @@ class UnetSkipConnectionBlock(nn.Module):
""" """
super(UnetSkipConnectionBlock, self).__init__() super(UnetSkipConnectionBlock, self).__init__()
self.outermost = outermost self.outermost = outermost
if type(norm_layer) == functools.partial: if isinstance(norm_layer, functools.partial):
use_bias = norm_layer.func == nn.InstanceNorm2d use_bias = norm_layer.func == nn.InstanceNorm2d
else: else:
use_bias = norm_layer == nn.InstanceNorm2d use_bias = norm_layer == nn.InstanceNorm2d

8
invokeai/backend/ip_adapter/ip_adapter.py
View File

@ -124,16 +124,14 @@ class IPAdapter(RawModel):
self.device, dtype=self.dtype self.device, dtype=self.dtype
) )
def to( def to(self, device: Optional[torch.device] = None, dtype: Optional[torch.dtype] = None):
self, device: Optional[torch.device] = None, dtype: Optional[torch.dtype] = None, non_blocking: bool = False
):
if device is not None: if device is not None:
self.device = device self.device = device
if dtype is not None: if dtype is not None:
self.dtype = dtype self.dtype = dtype
self._image_proj_model.to(device=self.device, dtype=self.dtype, non_blocking=non_blocking) self._image_proj_model.to(device=self.device, dtype=self.dtype)
self.attn_weights.to(device=self.device, dtype=self.dtype, non_blocking=non_blocking) self.attn_weights.to(device=self.device, dtype=self.dtype)
def calc_size(self) -> int: def calc_size(self) -> int:
# HACK(ryand): Fix this issue with circular imports. # HACK(ryand): Fix this issue with circular imports.

94
invokeai/backend/lora.py
View File

@ -11,7 +11,6 @@ from typing_extensions import Self
from invokeai.backend.model_manager import BaseModelType from invokeai.backend.model_manager import BaseModelType
from invokeai.backend.raw_model import RawModel from invokeai.backend.raw_model import RawModel
from invokeai.backend.util.devices import TorchDevice
class LoRALayerBase: class LoRALayerBase:
@ -57,14 +56,9 @@ class LoRALayerBase:
model_size += val.nelement() * val.element_size() model_size += val.nelement() * val.element_size()
return model_size return model_size
def to( def to(self, device: Optional[torch.device] = None, dtype: Optional[torch.dtype] = None) -> None:
self,
device: Optional[torch.device] = None,
dtype: Optional[torch.dtype] = None,
non_blocking: bool = False,
) -> None:
if self.bias is not None: if self.bias is not None:
self.bias = self.bias.to(device=device, dtype=dtype, non_blocking=non_blocking) self.bias = self.bias.to(device=device, dtype=dtype)
# TODO: find and debug lora/locon with bias # TODO: find and debug lora/locon with bias
@ -106,19 +100,14 @@ class LoRALayer(LoRALayerBase):
model_size += val.nelement() * val.element_size() model_size += val.nelement() * val.element_size()
return model_size return model_size
def to( def to(self, device: Optional[torch.device] = None, dtype: Optional[torch.dtype] = None) -> None:
self, super().to(device=device, dtype=dtype)
device: Optional[torch.device] = None,
dtype: Optional[torch.dtype] = None,
non_blocking: bool = False,
) -> None:
super().to(device=device, dtype=dtype, non_blocking=non_blocking)
self.up = self.up.to(device=device, dtype=dtype, non_blocking=non_blocking) self.up = self.up.to(device=device, dtype=dtype)
self.down = self.down.to(device=device, dtype=dtype, non_blocking=non_blocking) self.down = self.down.to(device=device, dtype=dtype)
if self.mid is not None: if self.mid is not None:
self.mid = self.mid.to(device=device, dtype=dtype, non_blocking=non_blocking) self.mid = self.mid.to(device=device, dtype=dtype)
class LoHALayer(LoRALayerBase): class LoHALayer(LoRALayerBase):
@ -167,23 +156,18 @@ class LoHALayer(LoRALayerBase):
model_size += val.nelement() * val.element_size() model_size += val.nelement() * val.element_size()
return model_size return model_size
def to( def to(self, device: Optional[torch.device] = None, dtype: Optional[torch.dtype] = None) -> None:
self,
device: Optional[torch.device] = None,
dtype: Optional[torch.dtype] = None,
non_blocking: bool = False,
) -> None:
super().to(device=device, dtype=dtype) super().to(device=device, dtype=dtype)
self.w1_a = self.w1_a.to(device=device, dtype=dtype, non_blocking=non_blocking) self.w1_a = self.w1_a.to(device=device, dtype=dtype)
self.w1_b = self.w1_b.to(device=device, dtype=dtype, non_blocking=non_blocking) self.w1_b = self.w1_b.to(device=device, dtype=dtype)
if self.t1 is not None: if self.t1 is not None:
self.t1 = self.t1.to(device=device, dtype=dtype, non_blocking=non_blocking) self.t1 = self.t1.to(device=device, dtype=dtype)
self.w2_a = self.w2_a.to(device=device, dtype=dtype, non_blocking=non_blocking) self.w2_a = self.w2_a.to(device=device, dtype=dtype)
self.w2_b = self.w2_b.to(device=device, dtype=dtype, non_blocking=non_blocking) self.w2_b = self.w2_b.to(device=device, dtype=dtype)
if self.t2 is not None: if self.t2 is not None:
self.t2 = self.t2.to(device=device, dtype=dtype, non_blocking=non_blocking) self.t2 = self.t2.to(device=device, dtype=dtype)
class LoKRLayer(LoRALayerBase): class LoKRLayer(LoRALayerBase):
@ -264,12 +248,7 @@ class LoKRLayer(LoRALayerBase):
model_size += val.nelement() * val.element_size() model_size += val.nelement() * val.element_size()
return model_size return model_size
def to( def to(self, device: Optional[torch.device] = None, dtype: Optional[torch.dtype] = None) -> None:
self,
device: Optional[torch.device] = None,
dtype: Optional[torch.dtype] = None,
non_blocking: bool = False,
) -> None:
super().to(device=device, dtype=dtype) super().to(device=device, dtype=dtype)
if self.w1 is not None: if self.w1 is not None:
@ -277,19 +256,19 @@ class LoKRLayer(LoRALayerBase):
else: else:
assert self.w1_a is not None assert self.w1_a is not None
assert self.w1_b is not None assert self.w1_b is not None
self.w1_a = self.w1_a.to(device=device, dtype=dtype, non_blocking=non_blocking) self.w1_a = self.w1_a.to(device=device, dtype=dtype)
self.w1_b = self.w1_b.to(device=device, dtype=dtype, non_blocking=non_blocking) self.w1_b = self.w1_b.to(device=device, dtype=dtype)
if self.w2 is not None: if self.w2 is not None:
self.w2 = self.w2.to(device=device, dtype=dtype, non_blocking=non_blocking) self.w2 = self.w2.to(device=device, dtype=dtype)
else: else:
assert self.w2_a is not None assert self.w2_a is not None
assert self.w2_b is not None assert self.w2_b is not None
self.w2_a = self.w2_a.to(device=device, dtype=dtype, non_blocking=non_blocking) self.w2_a = self.w2_a.to(device=device, dtype=dtype)
self.w2_b = self.w2_b.to(device=device, dtype=dtype, non_blocking=non_blocking) self.w2_b = self.w2_b.to(device=device, dtype=dtype)
if self.t2 is not None: if self.t2 is not None:
self.t2 = self.t2.to(device=device, dtype=dtype, non_blocking=non_blocking) self.t2 = self.t2.to(device=device, dtype=dtype)
class FullLayer(LoRALayerBase): class FullLayer(LoRALayerBase):
@ -319,15 +298,10 @@ class FullLayer(LoRALayerBase):
model_size += self.weight.nelement() * self.weight.element_size() model_size += self.weight.nelement() * self.weight.element_size()
return model_size return model_size
def to( def to(self, device: Optional[torch.device] = None, dtype: Optional[torch.dtype] = None) -> None:
self,
device: Optional[torch.device] = None,
dtype: Optional[torch.dtype] = None,
non_blocking: bool = False,
) -> None:
super().to(device=device, dtype=dtype) super().to(device=device, dtype=dtype)
self.weight = self.weight.to(device=device, dtype=dtype, non_blocking=non_blocking) self.weight = self.weight.to(device=device, dtype=dtype)
class IA3Layer(LoRALayerBase): class IA3Layer(LoRALayerBase):
@ -359,16 +333,11 @@ class IA3Layer(LoRALayerBase):
model_size += self.on_input.nelement() * self.on_input.element_size() model_size += self.on_input.nelement() * self.on_input.element_size()
return model_size return model_size
def to( def to(self, device: Optional[torch.device] = None, dtype: Optional[torch.dtype] = None):
self,
device: Optional[torch.device] = None,
dtype: Optional[torch.dtype] = None,
non_blocking: bool = False,
):
super().to(device=device, dtype=dtype) super().to(device=device, dtype=dtype)
self.weight = self.weight.to(device=device, dtype=dtype, non_blocking=non_blocking) self.weight = self.weight.to(device=device, dtype=dtype)
self.on_input = self.on_input.to(device=device, dtype=dtype, non_blocking=non_blocking) self.on_input = self.on_input.to(device=device, dtype=dtype)
AnyLoRALayer = Union[LoRALayer, LoHALayer, LoKRLayer, FullLayer, IA3Layer] AnyLoRALayer = Union[LoRALayer, LoHALayer, LoKRLayer, FullLayer, IA3Layer]
@ -390,15 +359,10 @@ class LoRAModelRaw(RawModel): # (torch.nn.Module):
def name(self) -> str: def name(self) -> str:
return self._name return self._name
def to( def to(self, device: Optional[torch.device] = None, dtype: Optional[torch.dtype] = None) -> None:
self,
device: Optional[torch.device] = None,
dtype: Optional[torch.dtype] = None,
non_blocking: bool = False,
) -> None:
# TODO: try revert if exception? # TODO: try revert if exception?
for _key, layer in self.layers.items(): for _key, layer in self.layers.items():
layer.to(device=device, dtype=dtype, non_blocking=non_blocking) layer.to(device=device, dtype=dtype)
def calc_size(self) -> int: def calc_size(self) -> int:
model_size = 0 model_size = 0
@ -521,7 +485,7 @@ class LoRAModelRaw(RawModel): # (torch.nn.Module):
# lower memory consumption by removing already parsed layer values # lower memory consumption by removing already parsed layer values
state_dict[layer_key].clear() state_dict[layer_key].clear()
layer.to(device=device, dtype=dtype, non_blocking=TorchDevice.get_non_blocking(device)) layer.to(device=device, dtype=dtype)
model.layers[layer_key] = layer model.layers[layer_key] = layer
return model return model

13
invokeai/backend/model_manager/config.py
View File

@ -67,6 +67,7 @@ class ModelType(str, Enum):
IPAdapter = "ip_adapter" IPAdapter = "ip_adapter"
CLIPVision = "clip_vision" CLIPVision = "clip_vision"
T2IAdapter = "t2i_adapter" T2IAdapter = "t2i_adapter"
SpandrelImageToImage = "spandrel_image_to_image"
class SubModelType(str, Enum): class SubModelType(str, Enum):
@ -371,6 +372,17 @@ class T2IAdapterConfig(DiffusersConfigBase, ControlAdapterConfigBase):
return Tag(f"{ModelType.T2IAdapter.value}.{ModelFormat.Diffusers.value}") return Tag(f"{ModelType.T2IAdapter.value}.{ModelFormat.Diffusers.value}")
class SpandrelImageToImageConfig(ModelConfigBase):
"""Model config for Spandrel Image to Image models."""
type: Literal[ModelType.SpandrelImageToImage] = ModelType.SpandrelImageToImage
format: Literal[ModelFormat.Checkpoint] = ModelFormat.Checkpoint
@staticmethod
def get_tag() -> Tag:
return Tag(f"{ModelType.SpandrelImageToImage.value}.{ModelFormat.Checkpoint.value}")
def get_model_discriminator_value(v: Any) -> str: def get_model_discriminator_value(v: Any) -> str:
""" """
Computes the discriminator value for a model config. Computes the discriminator value for a model config.
@ -407,6 +419,7 @@ AnyModelConfig = Annotated[
Annotated[IPAdapterInvokeAIConfig, IPAdapterInvokeAIConfig.get_tag()], Annotated[IPAdapterInvokeAIConfig, IPAdapterInvokeAIConfig.get_tag()],
Annotated[IPAdapterCheckpointConfig, IPAdapterCheckpointConfig.get_tag()], Annotated[IPAdapterCheckpointConfig, IPAdapterCheckpointConfig.get_tag()],
Annotated[T2IAdapterConfig, T2IAdapterConfig.get_tag()], Annotated[T2IAdapterConfig, T2IAdapterConfig.get_tag()],
Annotated[SpandrelImageToImageConfig, SpandrelImageToImageConfig.get_tag()],
Annotated[CLIPVisionDiffusersConfig, CLIPVisionDiffusersConfig.get_tag()], Annotated[CLIPVisionDiffusersConfig, CLIPVisionDiffusersConfig.get_tag()],
], ],
Discriminator(get_model_discriminator_value), Discriminator(get_model_discriminator_value),

6
invokeai/backend/model_manager/load/model_cache/model_cache_default.py
View File

@ -289,11 +289,9 @@ class ModelCache(ModelCacheBase[AnyModel]):
else: else:
new_dict: Dict[str, torch.Tensor] = {} new_dict: Dict[str, torch.Tensor] = {}
for k, v in cache_entry.state_dict.items(): for k, v in cache_entry.state_dict.items():
new_dict[k] = v.to( new_dict[k] = v.to(target_device, copy=True)
target_device, copy=True, non_blocking=TorchDevice.get_non_blocking(target_device)
)
cache_entry.model.load_state_dict(new_dict, assign=True) cache_entry.model.load_state_dict(new_dict, assign=True)
cache_entry.model.to(target_device, non_blocking=TorchDevice.get_non_blocking(target_device)) cache_entry.model.to(target_device)
cache_entry.device = target_device cache_entry.device = target_device
except Exception as e: # blow away cache entry except Exception as e: # blow away cache entry
self._delete_cache_entry(cache_entry) self._delete_cache_entry(cache_entry)

45
invokeai/backend/model_manager/load/model_loaders/spandrel_image_to_image.py Normal file
View File

@ -0,0 +1,45 @@
from pathlib import Path
from typing import Optional
import torch
from invokeai.backend.model_manager.config import (
AnyModel,
AnyModelConfig,
BaseModelType,
ModelFormat,
ModelType,
SubModelType,
)
from invokeai.backend.model_manager.load.load_default import ModelLoader
from invokeai.backend.model_manager.load.model_loader_registry import ModelLoaderRegistry
from invokeai.backend.spandrel_image_to_image_model import SpandrelImageToImageModel
@ModelLoaderRegistry.register(
base=BaseModelType.Any, type=ModelType.SpandrelImageToImage, format=ModelFormat.Checkpoint
)
class SpandrelImageToImageModelLoader(ModelLoader):
"""Class for loading Spandrel Image-to-Image models (i.e. models wrapped by spandrel.ImageModelDescriptor)."""
def _load_model(
self,
config: AnyModelConfig,
submodel_type: Optional[SubModelType] = None,
) -> AnyModel:
if submodel_type is not None:
raise ValueError("Unexpected submodel requested for Spandrel model.")
model_path = Path(config.path)
model = SpandrelImageToImageModel.load_from_file(model_path)
torch_dtype = self._torch_dtype
if not model.supports_dtype(torch_dtype):
self._logger.warning(
f"The configured dtype ('{self._torch_dtype}') is not supported by the {model.get_model_type_name()} "
"model. Falling back to 'float32'."
)
torch_dtype = torch.float32
model.to(dtype=torch_dtype)
return model

3
invokeai/backend/model_manager/load/model_util.py
View File

@ -15,6 +15,7 @@ from invokeai.backend.ip_adapter.ip_adapter import IPAdapter
from invokeai.backend.lora import LoRAModelRaw from invokeai.backend.lora import LoRAModelRaw
from invokeai.backend.model_manager.config import AnyModel from invokeai.backend.model_manager.config import AnyModel
from invokeai.backend.onnx.onnx_runtime import IAIOnnxRuntimeModel from invokeai.backend.onnx.onnx_runtime import IAIOnnxRuntimeModel
from invokeai.backend.spandrel_image_to_image_model import SpandrelImageToImageModel
from invokeai.backend.textual_inversion import TextualInversionModelRaw from invokeai.backend.textual_inversion import TextualInversionModelRaw
@ -33,7 +34,7 @@ def calc_model_size_by_data(logger: logging.Logger, model: AnyModel) -> int:
elif isinstance(model, CLIPTokenizer): elif isinstance(model, CLIPTokenizer):
# TODO(ryand): Accurately calculate the tokenizer's size. It's small enough that it shouldn't matter for now. # TODO(ryand): Accurately calculate the tokenizer's size. It's small enough that it shouldn't matter for now.
return 0 return 0
elif isinstance(model, (TextualInversionModelRaw, IPAdapter, LoRAModelRaw)): elif isinstance(model, (TextualInversionModelRaw, IPAdapter, LoRAModelRaw, SpandrelImageToImageModel)):
return model.calc_size() return model.calc_size()
else: else:
# TODO(ryand): Promote this from a log to an exception once we are confident that we are handling all of the # TODO(ryand): Promote this from a log to an exception once we are confident that we are handling all of the

56
invokeai/backend/model_manager/probe.py
View File

@ -4,6 +4,7 @@ from pathlib import Path
from typing import Any, Dict, Literal, Optional, Union from typing import Any, Dict, Literal, Optional, Union
import safetensors.torch import safetensors.torch
import spandrel
import torch import torch
from picklescan.scanner import scan_file_path from picklescan.scanner import scan_file_path
@ -25,6 +26,7 @@ from invokeai.backend.model_manager.config import (
SchedulerPredictionType, SchedulerPredictionType,
) )
from invokeai.backend.model_manager.util.model_util import lora_token_vector_length, read_checkpoint_meta from invokeai.backend.model_manager.util.model_util import lora_token_vector_length, read_checkpoint_meta
from invokeai.backend.spandrel_image_to_image_model import SpandrelImageToImageModel
from invokeai.backend.util.silence_warnings import SilenceWarnings from invokeai.backend.util.silence_warnings import SilenceWarnings
CkptType = Dict[str | int, Any] CkptType = Dict[str | int, Any]
@ -220,24 +222,46 @@ class ModelProbe(object):
ckpt = ckpt.get("state_dict", ckpt) ckpt = ckpt.get("state_dict", ckpt)
for key in [str(k) for k in ckpt.keys()]: for key in [str(k) for k in ckpt.keys()]:
if any(key.startswith(v) for v in {"cond_stage_model.", "first_stage_model.", "model.diffusion_model."}): if key.startswith(("cond_stage_model.", "first_stage_model.", "model.diffusion_model.")):
return ModelType.Main return ModelType.Main
elif any(key.startswith(v) for v in {"encoder.conv_in", "decoder.conv_in"}): elif key.startswith(("encoder.conv_in", "decoder.conv_in")):
return ModelType.VAE return ModelType.VAE
elif any(key.startswith(v) for v in {"lora_te_", "lora_unet_"}): elif key.startswith(("lora_te_", "lora_unet_")):
return ModelType.LoRA return ModelType.LoRA
elif any(key.endswith(v) for v in {"to_k_lora.up.weight", "to_q_lora.down.weight"}): elif key.endswith(("to_k_lora.up.weight", "to_q_lora.down.weight")):
return ModelType.LoRA return ModelType.LoRA
elif any(key.startswith(v) for v in {"controlnet", "control_model", "input_blocks"}): elif key.startswith(("controlnet", "control_model", "input_blocks")):
return ModelType.ControlNet return ModelType.ControlNet
elif any(key.startswith(v) for v in {"image_proj.", "ip_adapter."}): elif key.startswith(("image_proj.", "ip_adapter.")):
return ModelType.IPAdapter return ModelType.IPAdapter
elif key in {"emb_params", "string_to_param"}: elif key in {"emb_params", "string_to_param"}:
return ModelType.TextualInversion return ModelType.TextualInversion
else:
# diffusers-ti # diffusers-ti
if len(ckpt) < 10 and all(isinstance(v, torch.Tensor) for v in ckpt.values()): if len(ckpt) < 10 and all(isinstance(v, torch.Tensor) for v in ckpt.values()):
return ModelType.TextualInversion return ModelType.TextualInversion
# Check if the model can be loaded as a SpandrelImageToImageModel.
# This check is intentionally performed last, as it can be expensive (it requires loading the model from disk).
try:
# It would be nice to avoid having to load the Spandrel model from disk here. A couple of options were
# explored to avoid this:
# 1. Call `SpandrelImageToImageModel.load_from_state_dict(ckpt)`, where `ckpt` is a state_dict on the meta
# device. Unfortunately, some Spandrel models perform operations during initialization that are not
# supported on meta tensors.
# 2. Spandrel has internal logic to determine a model's type from its state_dict before loading the model.
# This logic is not exposed in spandrel's public API. We could copy the logic here, but then we have to
# maintain it, and the risk of false positive detections is higher.
SpandrelImageToImageModel.load_from_file(model_path)
return ModelType.SpandrelImageToImage
except spandrel.UnsupportedModelError:
pass
except RuntimeError as e:
if "No such file or directory" in str(e):
# This error is expected if the model_path does not exist (which is the case in some unit tests).
pass
else:
raise e
raise InvalidModelConfigException(f"Unable to determine model type for {model_path}") raise InvalidModelConfigException(f"Unable to determine model type for {model_path}")
@ -569,6 +593,11 @@ class T2IAdapterCheckpointProbe(CheckpointProbeBase):
raise NotImplementedError() raise NotImplementedError()
class SpandrelImageToImageCheckpointProbe(CheckpointProbeBase):
def get_base_type(self) -> BaseModelType:
return BaseModelType.Any
######################################################## ########################################################
# classes for probing folders # classes for probing folders
####################################################### #######################################################
@ -776,6 +805,11 @@ class CLIPVisionFolderProbe(FolderProbeBase):
return BaseModelType.Any return BaseModelType.Any
class SpandrelImageToImageFolderProbe(FolderProbeBase):
def get_base_type(self) -> BaseModelType:
raise NotImplementedError()
class T2IAdapterFolderProbe(FolderProbeBase): class T2IAdapterFolderProbe(FolderProbeBase):
def get_base_type(self) -> BaseModelType: def get_base_type(self) -> BaseModelType:
config_file = self.model_path / "config.json" config_file = self.model_path / "config.json"
@ -805,6 +839,7 @@ ModelProbe.register_probe("diffusers", ModelType.ControlNet, ControlNetFolderPro
ModelProbe.register_probe("diffusers", ModelType.IPAdapter, IPAdapterFolderProbe) ModelProbe.register_probe("diffusers", ModelType.IPAdapter, IPAdapterFolderProbe)
ModelProbe.register_probe("diffusers", ModelType.CLIPVision, CLIPVisionFolderProbe) ModelProbe.register_probe("diffusers", ModelType.CLIPVision, CLIPVisionFolderProbe)
ModelProbe.register_probe("diffusers", ModelType.T2IAdapter, T2IAdapterFolderProbe) ModelProbe.register_probe("diffusers", ModelType.T2IAdapter, T2IAdapterFolderProbe)
ModelProbe.register_probe("diffusers", ModelType.SpandrelImageToImage, SpandrelImageToImageFolderProbe)
ModelProbe.register_probe("checkpoint", ModelType.Main, PipelineCheckpointProbe) ModelProbe.register_probe("checkpoint", ModelType.Main, PipelineCheckpointProbe)
ModelProbe.register_probe("checkpoint", ModelType.VAE, VaeCheckpointProbe) ModelProbe.register_probe("checkpoint", ModelType.VAE, VaeCheckpointProbe)
@ -814,5 +849,6 @@ ModelProbe.register_probe("checkpoint", ModelType.ControlNet, ControlNetCheckpoi
ModelProbe.register_probe("checkpoint", ModelType.IPAdapter, IPAdapterCheckpointProbe) ModelProbe.register_probe("checkpoint", ModelType.IPAdapter, IPAdapterCheckpointProbe)
ModelProbe.register_probe("checkpoint", ModelType.CLIPVision, CLIPVisionCheckpointProbe) ModelProbe.register_probe("checkpoint", ModelType.CLIPVision, CLIPVisionCheckpointProbe)
ModelProbe.register_probe("checkpoint", ModelType.T2IAdapter, T2IAdapterCheckpointProbe) ModelProbe.register_probe("checkpoint", ModelType.T2IAdapter, T2IAdapterCheckpointProbe)
ModelProbe.register_probe("checkpoint", ModelType.SpandrelImageToImage, SpandrelImageToImageCheckpointProbe)
ModelProbe.register_probe("onnx", ModelType.ONNX, ONNXFolderProbe) ModelProbe.register_probe("onnx", ModelType.ONNX, ONNXFolderProbe)

37
invokeai/backend/model_manager/starter_models.py
View File

@ -399,6 +399,43 @@ STARTER_MODELS: list[StarterModel] = [
type=ModelType.T2IAdapter, type=ModelType.T2IAdapter,
), ),
# endregion # endregion
# region SpandrelImageToImage
StarterModel(
name="RealESRGAN_x4plus_anime_6B",
base=BaseModelType.Any,
source="https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.2.4/RealESRGAN_x4plus_anime_6B.pth",
description="A Real-ESRGAN 4x upscaling model (optimized for anime images).",
type=ModelType.SpandrelImageToImage,
),
StarterModel(
name="RealESRGAN_x4plus",
base=BaseModelType.Any,
source="https://github.com/xinntao/Real-ESRGAN/releases/download/v0.1.0/RealESRGAN_x4plus.pth",
description="A Real-ESRGAN 4x upscaling model (general-purpose).",
type=ModelType.SpandrelImageToImage,
),
StarterModel(
name="ESRGAN_SRx4_DF2KOST_official",
base=BaseModelType.Any,
source="https://github.com/xinntao/Real-ESRGAN/releases/download/v0.1.1/ESRGAN_SRx4_DF2KOST_official-ff704c30.pth",
description="The official ESRGAN 4x upscaling model.",
type=ModelType.SpandrelImageToImage,
),
StarterModel(
name="RealESRGAN_x2plus",
base=BaseModelType.Any,
source="https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.1/RealESRGAN_x2plus.pth",
description="A Real-ESRGAN 2x upscaling model (general-purpose).",
type=ModelType.SpandrelImageToImage,
),
StarterModel(
name="SwinIR - realSR_BSRGAN_DFOWMFC_s64w8_SwinIR-L_x4_GAN",
base=BaseModelType.Any,
source="https://github.com/JingyunLiang/SwinIR/releases/download/v0.0/003_realSR_BSRGAN_DFOWMFC_s64w8_SwinIR-L_x4_GAN-with-dict-keys-params-and-params_ema.pth",
description="A SwinIR 4x upscaling model.",
type=ModelType.SpandrelImageToImage,
),
# endregion
] ]
assert len(STARTER_MODELS) == len({m.source for m in STARTER_MODELS}), "Duplicate starter models" assert len(STARTER_MODELS) == len({m.source for m in STARTER_MODELS}), "Duplicate starter models"

15
invokeai/backend/model_patcher.py
View File

@ -158,15 +158,12 @@ class ModelPatcher:
# We intentionally move to the target device first, then cast. Experimentally, this was found to # We intentionally move to the target device first, then cast. Experimentally, this was found to
# be significantly faster for 16-bit CPU tensors being moved to a CUDA device than doing the # be significantly faster for 16-bit CPU tensors being moved to a CUDA device than doing the
# same thing in a single call to '.to(...)'. # same thing in a single call to '.to(...)'.
layer.to(device=device, non_blocking=TorchDevice.get_non_blocking(device)) layer.to(device=device)
layer.to(dtype=torch.float32, non_blocking=TorchDevice.get_non_blocking(device)) layer.to(dtype=torch.float32)
# TODO(ryand): Using torch.autocast(...) over explicit casting may offer a speed benefit on CUDA # TODO(ryand): Using torch.autocast(...) over explicit casting may offer a speed benefit on CUDA
# devices here. Experimentally, it was found to be very slow on CPU. More investigation needed. # devices here. Experimentally, it was found to be very slow on CPU. More investigation needed.
layer_weight = layer.get_weight(module.weight) * (lora_weight * layer_scale) layer_weight = layer.get_weight(module.weight) * (lora_weight * layer_scale)
layer.to( layer.to(device=TorchDevice.CPU_DEVICE)
device=TorchDevice.CPU_DEVICE,
non_blocking=TorchDevice.get_non_blocking(TorchDevice.CPU_DEVICE),
)
assert isinstance(layer_weight, torch.Tensor) # mypy thinks layer_weight is a float|Any ??! assert isinstance(layer_weight, torch.Tensor) # mypy thinks layer_weight is a float|Any ??!
if module.weight.shape != layer_weight.shape: if module.weight.shape != layer_weight.shape:
@ -175,7 +172,7 @@ class ModelPatcher:
layer_weight = layer_weight.reshape(module.weight.shape) layer_weight = layer_weight.reshape(module.weight.shape)
assert isinstance(layer_weight, torch.Tensor) # mypy thinks layer_weight is a float|Any ??! assert isinstance(layer_weight, torch.Tensor) # mypy thinks layer_weight is a float|Any ??!
module.weight += layer_weight.to(dtype=dtype, non_blocking=TorchDevice.get_non_blocking(device)) module.weight += layer_weight.to(dtype=dtype)
yield # wait for context manager exit yield # wait for context manager exit
@ -183,9 +180,7 @@ class ModelPatcher:
assert hasattr(model, "get_submodule") # mypy not picking up fact that torch.nn.Module has get_submodule() assert hasattr(model, "get_submodule") # mypy not picking up fact that torch.nn.Module has get_submodule()
with torch.no_grad(): with torch.no_grad():
for module_key, weight in original_weights.items(): for module_key, weight in original_weights.items():
model.get_submodule(module_key).weight.copy_( model.get_submodule(module_key).weight.copy_(weight)
weight, non_blocking=TorchDevice.get_non_blocking(weight.device)
)
@classmethod @classmethod
@contextmanager @contextmanager

7
invokeai/backend/onnx/onnx_runtime.py
View File

@ -190,12 +190,7 @@ class IAIOnnxRuntimeModel(RawModel):
return self.session.run(None, inputs) return self.session.run(None, inputs)
# compatability with RawModel ABC # compatability with RawModel ABC
def to( def to(self, device: Optional[torch.device] = None, dtype: Optional[torch.dtype] = None) -> None:
self,
device: Optional[torch.device] = None,
dtype: Optional[torch.dtype] = None,
non_blocking: bool = False,
) -> None:
pass pass
# compatability with diffusers load code # compatability with diffusers load code

31
invokeai/backend/raw_model.py
View File

@ -1,15 +1,3 @@
"""Base class for 'Raw' models.
The RawModel class is the base class of LoRAModelRaw and TextualInversionModelRaw,
and is used for type checking of calls to the model patcher. Its main purpose
is to avoid a circular import issues when lora.py tries to import BaseModelType
from invokeai.backend.model_manager.config, and the latter tries to import LoRAModelRaw
from lora.py.
The term 'raw' was introduced to describe a wrapper around a torch.nn.Module
that adds additional methods and attributes.
"""
from abc import ABC, abstractmethod from abc import ABC, abstractmethod
from typing import Optional from typing import Optional
@ -17,13 +5,18 @@ import torch
class RawModel(ABC): class RawModel(ABC):
"""Abstract base class for 'Raw' model wrappers.""" """Base class for 'Raw' models.
The RawModel class is the base class of LoRAModelRaw, TextualInversionModelRaw, etc.
and is used for type checking of calls to the model patcher. Its main purpose
is to avoid a circular import issues when lora.py tries to import BaseModelType
from invokeai.backend.model_manager.config, and the latter tries to import LoRAModelRaw
from lora.py.
The term 'raw' was introduced to describe a wrapper around a torch.nn.Module
that adds additional methods and attributes.
"""
@abstractmethod @abstractmethod
def to( def to(self, device: Optional[torch.device] = None, dtype: Optional[torch.dtype] = None) -> None:
self,
device: Optional[torch.device] = None,
dtype: Optional[torch.dtype] = None,
non_blocking: bool = False,
) -> None:
pass pass

139
invokeai/backend/spandrel_image_to_image_model.py Normal file
View File

@ -0,0 +1,139 @@
from pathlib import Path
from typing import Any, Optional
import numpy as np
import torch
from PIL import Image
from spandrel import ImageModelDescriptor, ModelLoader
from invokeai.backend.raw_model import RawModel
class SpandrelImageToImageModel(RawModel):
"""A wrapper for a Spandrel Image-to-Image model.
The main reason for having a wrapper class is to integrate with the type handling of RawModel.
"""
def __init__(self, spandrel_model: ImageModelDescriptor[Any]):
self._spandrel_model = spandrel_model
@staticmethod
def pil_to_tensor(image: Image.Image) -> torch.Tensor:
"""Convert PIL Image to the torch.Tensor format expected by SpandrelImageToImageModel.run().
Args:
image (Image.Image): A PIL Image with shape (H, W, C) and values in the range [0, 255].
Returns:
torch.Tensor: A torch.Tensor with shape (N, C, H, W) and values in the range [0, 1].
"""
image_np = np.array(image)
# (H, W, C) -> (C, H, W)
image_np = np.transpose(image_np, (2, 0, 1))
image_np = image_np / 255
image_tensor = torch.from_numpy(image_np).float()
# (C, H, W) -> (N, C, H, W)
image_tensor = image_tensor.unsqueeze(0)
return image_tensor
@staticmethod
def tensor_to_pil(tensor: torch.Tensor) -> Image.Image:
"""Convert a torch.Tensor produced by SpandrelImageToImageModel.run() to a PIL Image.
Args:
tensor (torch.Tensor): A torch.Tensor with shape (N, C, H, W) and values in the range [0, 1].
Returns:
Image.Image: A PIL Image with shape (H, W, C) and values in the range [0, 255].
"""
# (N, C, H, W) -> (C, H, W)
tensor = tensor.squeeze(0)
# (C, H, W) -> (H, W, C)
tensor = tensor.permute(1, 2, 0)
tensor = tensor.clamp(0, 1)
tensor = (tensor * 255).cpu().detach().numpy().astype(np.uint8)
image = Image.fromarray(tensor)
return image
def run(self, image_tensor: torch.Tensor) -> torch.Tensor:
"""Run the image-to-image model.
Args:
image_tensor (torch.Tensor): A torch.Tensor with shape (N, C, H, W) and values in the range [0, 1].
"""
return self._spandrel_model(image_tensor)
@classmethod
def load_from_file(cls, file_path: str | Path):
model = ModelLoader().load_from_file(file_path)
if not isinstance(model, ImageModelDescriptor):
raise ValueError(
f"Loaded a spandrel model of type '{type(model)}'. Only image-to-image models are supported "
"('ImageModelDescriptor')."
)
return cls(spandrel_model=model)
@classmethod
def load_from_state_dict(cls, state_dict: dict[str, torch.Tensor]):
model = ModelLoader().load_from_state_dict(state_dict)
if not isinstance(model, ImageModelDescriptor):
raise ValueError(
f"Loaded a spandrel model of type '{type(model)}'. Only image-to-image models are supported "
"('ImageModelDescriptor')."
)
return cls(spandrel_model=model)
def supports_dtype(self, dtype: torch.dtype) -> bool:
"""Check if the model supports the given dtype."""
if dtype == torch.float16:
return self._spandrel_model.supports_half
elif dtype == torch.bfloat16:
return self._spandrel_model.supports_bfloat16
elif dtype == torch.float32:
# All models support float32.
return True
else:
raise ValueError(f"Unexpected dtype '{dtype}'.")
def get_model_type_name(self) -> str:
"""The model type name. Intended for logging / debugging purposes. Do not rely on this field remaining
consistent over time.
"""
return str(type(self._spandrel_model.model))
def to(
self,
device: Optional[torch.device] = None,
dtype: Optional[torch.dtype] = None,
non_blocking: bool = False,
) -> None:
"""Note: Some models have limited dtype support. Call supports_dtype(...) to check if the dtype is supported.
Note: The non_blocking parameter is currently ignored."""
# TODO(ryand): spandrel.ImageModelDescriptor.to(...) does not support non_blocking. We will have to access the
# model directly if we want to apply this optimization.
self._spandrel_model.to(device=device, dtype=dtype)
@property
def device(self) -> torch.device:
"""The device of the underlying model."""
return self._spandrel_model.device
@property
def dtype(self) -> torch.dtype:
"""The dtype of the underlying model."""
return self._spandrel_model.dtype
@property
def scale(self) -> int:
"""The scale of the model (e.g. 1x, 2x, 4x, etc.)."""
return self._spandrel_model.scale
def calc_size(self) -> int:
"""Get size of the model in memory in bytes."""
# HACK(ryand): Fix this issue with circular imports.
from invokeai.backend.model_manager.load.model_util import calc_module_size
return calc_module_size(self._spandrel_model.model)

9
invokeai/backend/textual_inversion.py
View File

@ -65,17 +65,12 @@ class TextualInversionModelRaw(RawModel):
return result return result
def to( def to(self, device: Optional[torch.device] = None, dtype: Optional[torch.dtype] = None) -> None:
self,
device: Optional[torch.device] = None,
dtype: Optional[torch.dtype] = None,
non_blocking: bool = False,
) -> None:
if not torch.cuda.is_available(): if not torch.cuda.is_available():
return return
for emb in [self.embedding, self.embedding_2]: for emb in [self.embedding, self.embedding_2]:
if emb is not None: if emb is not None:
emb.to(device=device, dtype=dtype, non_blocking=non_blocking) emb.to(device=device, dtype=dtype)
def calc_size(self) -> int: def calc_size(self) -> int:
"""Get the size of this model in bytes.""" """Get the size of this model in bytes."""

12
invokeai/backend/util/devices.py
View File

@ -112,15 +112,3 @@ class TorchDevice:
@classmethod @classmethod
def _to_dtype(cls, precision_name: TorchPrecisionNames) -> torch.dtype: def _to_dtype(cls, precision_name: TorchPrecisionNames) -> torch.dtype:
return NAME_TO_PRECISION[precision_name] return NAME_TO_PRECISION[precision_name]
@staticmethod
def get_non_blocking(to_device: torch.device) -> bool:
"""Return the non_blocking flag to be used when moving a tensor to a given device.
MPS may have unexpected errors with non-blocking operations - we should not use non-blocking when moving _to_ MPS.
When moving _from_ MPS, we can use non-blocking operations.
See:
- https://github.com/pytorch/pytorch/issues/107455
- https://discuss.pytorch.org/t/should-we-set-non-blocking-to-true/38234/28
"""
return False if to_device.type == "mps" else True

8
invokeai/frontend/web/public/locales/it.json
View File

@ -962,8 +962,8 @@
"selectedForAutoAdd": "Selezionato per l'aggiunta automatica", "selectedForAutoAdd": "Selezionato per l'aggiunta automatica",
"addSharedBoard": "Aggiungi una Bacheca Condivisa", "addSharedBoard": "Aggiungi una Bacheca Condivisa",
"boards": "Bacheche", "boards": "Bacheche",
"private": "Privata", "private": "Bacheche private",
"shared": "Condivisa", "shared": "Bacheche condivise",
"addPrivateBoard": "Aggiungi una Bacheca Privata" "addPrivateBoard": "Aggiungi una Bacheca Privata"
}, },
"controlnet": { "controlnet": {
@ -1028,7 +1028,7 @@
"minConfidence": "Confidenza minima", "minConfidence": "Confidenza minima",
"scribble": "Scarabocchio", "scribble": "Scarabocchio",
"amult": "Angolo di illuminazione", "amult": "Angolo di illuminazione",
"coarse": "Approssimativo", "coarse": "Grossolano",
"resizeSimple": "Ridimensiona (semplice)", "resizeSimple": "Ridimensiona (semplice)",
"large": "Grande", "large": "Grande",
"small": "Piccolo", "small": "Piccolo",
@ -1353,7 +1353,7 @@
"lora": { "lora": {
"heading": "LoRA", "heading": "LoRA",
"paragraphs": [ "paragraphs": [
"Modelli leggeri utilizzati insieme ai modelli base." "Modelli concettuali utilizzati insieme ai modelli di base."
] ]
}, },
"controlNet": { "controlNet": {

9
invokeai/frontend/web/src/app/store/middleware/listenerMiddleware/listeners/imageDeletionListeners.ts
View File

@ -136,7 +136,12 @@ export const addImageDeletionListeners = (startAppListening: AppStartListening)
if (data) { if (data) {
const deletedImageIndex = data.items.findIndex((i) => i.image_name === imageDTO.image_name); const deletedImageIndex = data.items.findIndex((i) => i.image_name === imageDTO.image_name);
const nextImage = data.items[deletedImageIndex + 1] ?? data.items[0] ?? null; const nextImage = data.items[deletedImageIndex + 1] ?? data.items[0] ?? null;
dispatch(imageSelected(nextImage)); if (nextImage?.image_name === imageDTO.image_name) {
// If the next image is the same as the deleted one, it means it was the last image, reset selection
dispatch(imageSelected(null));
} else {
dispatch(imageSelected(nextImage));
}
} }
} }
@ -176,6 +181,8 @@ export const addImageDeletionListeners = (startAppListening: AppStartListening)
const queryArgs = selectListImagesQueryArgs(state); const queryArgs = selectListImagesQueryArgs(state);
const { data } = imagesApi.endpoints.listImages.select(queryArgs)(state); const { data } = imagesApi.endpoints.listImages.select(queryArgs)(state);
if (data) { if (data) {
// When we delete multiple images, we clear the selection. Then, the the next time we load images, we will
// select the first one. This is handled below in the listener for `imagesApi.endpoints.listImages.matchFulfilled`.
dispatch(imageSelected(null)); dispatch(imageSelected(null));
} }
} }

86
invokeai/frontend/web/src/features/gallery/components/Boards/BoardsList/BoardsList.tsx
View File

@ -1,4 +1,4 @@
import { Flex, Text } from '@invoke-ai/ui-library'; import { Box, Flex, Text } from '@invoke-ai/ui-library';
import { EMPTY_ARRAY } from 'app/store/constants'; import { EMPTY_ARRAY } from 'app/store/constants';
import { useAppSelector } from 'app/store/storeHooks'; import { useAppSelector } from 'app/store/storeHooks';
import { overlayScrollbarsParams } from 'common/components/OverlayScrollbars/constants'; import { overlayScrollbarsParams } from 'common/components/OverlayScrollbars/constants';
@ -40,9 +40,41 @@ const BoardsList = () => {
return ( return (
<> <>
<Flex flexDir="column" gap={2} borderRadius="base" maxHeight="100%"> <Box position="relative" w="full" h="full">
<OverlayScrollbarsComponent defer style={overlayScrollbarsStyles} options={overlayScrollbarsParams.options}> <Box position="absolute" top={0} right={0} bottom={0} left={0}>
{allowPrivateBoards && ( <OverlayScrollbarsComponent defer style={overlayScrollbarsStyles} options={overlayScrollbarsParams.options}>
{allowPrivateBoards && (
<Flex direction="column" gap={1}>
<Flex
position="sticky"
w="full"
justifyContent="space-between"
alignItems="center"
ps={2}
pb={1}
pt={2}
zIndex={1}
top={0}
bg="base.900"
>
<Text fontSize="md" fontWeight="semibold" userSelect="none">
{t('boards.private')}
</Text>
<AddBoardButton isPrivateBoard={true} />
</Flex>
<Flex direction="column" gap={1}>
<NoBoardBoard isSelected={selectedBoardId === 'none'} />
{filteredPrivateBoards.map((board) => (
<GalleryBoard
board={board}
isSelected={selectedBoardId === board.board_id}
setBoardToDelete={setBoardToDelete}
key={board.board_id}
/>
))}
</Flex>
</Flex>
)}
<Flex direction="column" gap={1}> <Flex direction="column" gap={1}>
<Flex <Flex
position="sticky" position="sticky"
@ -50,19 +82,20 @@ const BoardsList = () => {
justifyContent="space-between" justifyContent="space-between"
alignItems="center" alignItems="center"
ps={2} ps={2}
py={1} pb={1}
pt={2}
zIndex={1} zIndex={1}
top={0} top={0}
bg="base.900" bg="base.900"
> >
<Text fontSize="md" fontWeight="semibold" userSelect="none"> <Text fontSize="md" fontWeight="semibold" userSelect="none">
{t('boards.private')} {allowPrivateBoards ? t('boards.shared') : t('boards.boards')}
</Text> </Text>
<AddBoardButton isPrivateBoard={true} /> <AddBoardButton isPrivateBoard={false} />
</Flex> </Flex>
<Flex direction="column" gap={1}> <Flex direction="column" gap={1}>
<NoBoardBoard isSelected={selectedBoardId === 'none'} /> {!allowPrivateBoards && <NoBoardBoard isSelected={selectedBoardId === 'none'} />}
{filteredPrivateBoards.map((board) => ( {filteredSharedBoards.map((board) => (
<GalleryBoard <GalleryBoard
board={board} board={board}
isSelected={selectedBoardId === board.board_id} isSelected={selectedBoardId === board.board_id}
@ -72,38 +105,9 @@ const BoardsList = () => {
))} ))}
</Flex> </Flex>
</Flex> </Flex>
)} </OverlayScrollbarsComponent>
<Flex direction="column" gap={1} pb={2}> </Box>
<Flex </Box>
position="sticky"
w="full"
justifyContent="space-between"
alignItems="center"
ps={2}
py={1}
zIndex={1}
top={0}
bg="base.900"
>
<Text fontSize="md" fontWeight="semibold" userSelect="none">
{allowPrivateBoards ? t('boards.shared') : t('boards.boards')}
</Text>
<AddBoardButton isPrivateBoard={false} />
</Flex>
<Flex direction="column" gap={1}>
{!allowPrivateBoards && <NoBoardBoard isSelected={selectedBoardId === 'none'} />}
{filteredSharedBoards.map((board) => (
<GalleryBoard
board={board}
isSelected={selectedBoardId === board.board_id}
setBoardToDelete={setBoardToDelete}
key={board.board_id}
/>
))}
</Flex>
</Flex>
</OverlayScrollbarsComponent>
</Flex>
<DeleteBoardModal boardToDelete={boardToDelete} setBoardToDelete={setBoardToDelete} /> <DeleteBoardModal boardToDelete={boardToDelete} setBoardToDelete={setBoardToDelete} />
</> </>
); );

25
invokeai/frontend/web/src/features/gallery/components/ImageGalleryContent.tsx
View File

@ -16,6 +16,7 @@ import { GalleryHeader } from 'features/gallery/components/GalleryHeader';
import { galleryViewChanged } from 'features/gallery/store/gallerySlice'; import { galleryViewChanged } from 'features/gallery/store/gallerySlice';
import ResizeHandle from 'features/ui/components/tabs/ResizeHandle'; import ResizeHandle from 'features/ui/components/tabs/ResizeHandle';
import { usePanel, type UsePanelOptions } from 'features/ui/hooks/usePanel'; import { usePanel, type UsePanelOptions } from 'features/ui/hooks/usePanel';
import type { CSSProperties } from 'react';
import { memo, useCallback, useMemo, useRef } from 'react'; import { memo, useCallback, useMemo, useRef } from 'react';
import { useTranslation } from 'react-i18next'; import { useTranslation } from 'react-i18next';
import { PiMagnifyingGlassBold } from 'react-icons/pi'; import { PiMagnifyingGlassBold } from 'react-icons/pi';
@ -29,13 +30,15 @@ import GalleryImageGrid from './ImageGrid/GalleryImageGrid';
import { GalleryPagination } from './ImageGrid/GalleryPagination'; import { GalleryPagination } from './ImageGrid/GalleryPagination';
import { GallerySearch } from './ImageGrid/GallerySearch'; import { GallerySearch } from './ImageGrid/GallerySearch';
const baseStyles: ChakraProps['sx'] = { const COLLAPSE_STYLES: CSSProperties = { flexShrink: 0, minHeight: 0 };
const BASE_STYLES: ChakraProps['sx'] = {
fontWeight: 'semibold', fontWeight: 'semibold',
fontSize: 'sm', fontSize: 'sm',
color: 'base.300', color: 'base.300',
}; };
const selectedStyles: ChakraProps['sx'] = { const SELECTED_STYLES: ChakraProps['sx'] = {
borderColor: 'base.800', borderColor: 'base.800',
borderBottomColor: 'base.900', borderBottomColor: 'base.900',
color: 'invokeBlue.300', color: 'invokeBlue.300',
@ -110,11 +113,13 @@ const ImageGalleryContent = () => {
onExpand={boardsListPanel.onExpand} onExpand={boardsListPanel.onExpand}
collapsible collapsible
> >
<Collapse in={boardSearchDisclosure.isOpen}> <Flex flexDir="column" w="full" h="full">
<BoardsSearch /> <Collapse in={boardSearchDisclosure.isOpen} style={COLLAPSE_STYLES}>
</Collapse> <BoardsSearch />
<Divider pt={2} /> </Collapse>
<BoardsList /> <Divider pt={2} />
<BoardsList />
</Flex>
</Panel> </Panel>
<ResizeHandle <ResizeHandle
id="gallery-panel-handle" id="gallery-panel-handle"
@ -125,10 +130,10 @@ const ImageGalleryContent = () => {
<Flex flexDirection="column" alignItems="center" justifyContent="space-between" h="full" w="full"> <Flex flexDirection="column" alignItems="center" justifyContent="space-between" h="full" w="full">
<Tabs index={galleryView === 'images' ? 0 : 1} variant="enclosed" display="flex" flexDir="column" w="full"> <Tabs index={galleryView === 'images' ? 0 : 1} variant="enclosed" display="flex" flexDir="column" w="full">
<TabList gap={2} fontSize="sm" borderColor="base.800"> <TabList gap={2} fontSize="sm" borderColor="base.800">
<Tab sx={baseStyles} _selected={selectedStyles} onClick={handleClickImages} data-testid="images-tab"> <Tab sx={BASE_STYLES} _selected={SELECTED_STYLES} onClick={handleClickImages} data-testid="images-tab">
{t('parameters.images')} {t('parameters.images')}
</Tab> </Tab>
<Tab sx={baseStyles} _selected={selectedStyles} onClick={handleClickAssets} data-testid="assets-tab"> <Tab sx={BASE_STYLES} _selected={SELECTED_STYLES} onClick={handleClickAssets} data-testid="assets-tab">
{t('gallery.assets')} {t('gallery.assets')}
</Tab> </Tab>
<Spacer /> <Spacer />
@ -157,7 +162,7 @@ const ImageGalleryContent = () => {
</TabList> </TabList>
</Tabs> </Tabs>
<Box w="full"> <Box w="full">
<Collapse in={searchDisclosure.isOpen}> <Collapse in={searchDisclosure.isOpen} style={COLLAPSE_STYLES}>
<Box w="full" pt={2}> <Box w="full" pt={2}>
<GallerySearch /> <GallerySearch />
</Box> </Box>

23
invokeai/frontend/web/src/features/modelManagerV2/subpanels/ModelManagerPanel/ModelList.tsx
View File

@ -11,6 +11,7 @@ import {
useLoRAModels, useLoRAModels,
useMainModels, useMainModels,
useRefinerModels, useRefinerModels,
useSpandrelImageToImageModels,
useT2IAdapterModels, useT2IAdapterModels,
useVAEModels, useVAEModels,
} from 'services/api/hooks/modelsByType'; } from 'services/api/hooks/modelsByType';
@ -71,6 +72,13 @@ const ModelList = () => {
[vaeModels, searchTerm, filteredModelType] [vaeModels, searchTerm, filteredModelType]
); );
const [spandrelImageToImageModels, { isLoading: isLoadingSpandrelImageToImageModels }] =
useSpandrelImageToImageModels();
const filteredSpandrelImageToImageModels = useMemo(
() => modelsFilter(spandrelImageToImageModels, searchTerm, filteredModelType),
[spandrelImageToImageModels, searchTerm, filteredModelType]
);
const totalFilteredModels = useMemo(() => { const totalFilteredModels = useMemo(() => {
return ( return (
filteredMainModels.length + filteredMainModels.length +
@ -80,7 +88,8 @@ const ModelList = () => {
filteredControlNetModels.length + filteredControlNetModels.length +
filteredT2IAdapterModels.length + filteredT2IAdapterModels.length +
filteredIPAdapterModels.length + filteredIPAdapterModels.length +
filteredVAEModels.length filteredVAEModels.length +
filteredSpandrelImageToImageModels.length
); );
}, [ }, [
filteredControlNetModels.length, filteredControlNetModels.length,
@ -91,6 +100,7 @@ const ModelList = () => {
filteredRefinerModels.length, filteredRefinerModels.length,
filteredT2IAdapterModels.length, filteredT2IAdapterModels.length,
filteredVAEModels.length, filteredVAEModels.length,
filteredSpandrelImageToImageModels.length,
]); ]);
return ( return (
@ -143,6 +153,17 @@ const ModelList = () => {
{!isLoadingT2IAdapterModels && filteredT2IAdapterModels.length > 0 && ( {!isLoadingT2IAdapterModels && filteredT2IAdapterModels.length > 0 && (
<ModelListWrapper title={t('common.t2iAdapter')} modelList={filteredT2IAdapterModels} key="t2i-adapters" /> <ModelListWrapper title={t('common.t2iAdapter')} modelList={filteredT2IAdapterModels} key="t2i-adapters" />
)} )}
{/* Spandrel Image to Image List */}
{isLoadingSpandrelImageToImageModels && (
<FetchingModelsLoader loadingMessage="Loading Image-to-Image Models..." />
)}
{!isLoadingSpandrelImageToImageModels && filteredSpandrelImageToImageModels.length > 0 && (
<ModelListWrapper
title="Image-to-Image"
modelList={filteredSpandrelImageToImageModels}
key="spandrel-image-to-image"
/>
)}
{totalFilteredModels === 0 && ( {totalFilteredModels === 0 && (
<Flex w="full" h="full" alignItems="center" justifyContent="center"> <Flex w="full" h="full" alignItems="center" justifyContent="center">
<Text>{t('modelManager.noMatchingModels')}</Text> <Text>{t('modelManager.noMatchingModels')}</Text>

1
invokeai/frontend/web/src/features/modelManagerV2/subpanels/ModelManagerPanel/ModelTypeFilter.tsx
View File

@ -21,6 +21,7 @@ export const ModelTypeFilter = () => {
t2i_adapter: t('common.t2iAdapter'), t2i_adapter: t('common.t2iAdapter'),
ip_adapter: t('common.ipAdapter'), ip_adapter: t('common.ipAdapter'),
clip_vision: 'Clip Vision', clip_vision: 'Clip Vision',
spandrel_image_to_image: 'Image-to-Image',
}), }),
[t] [t]
); );

17
invokeai/frontend/web/src/features/nodes/components/flow/nodes/Invocation/fields/InputFieldRenderer.tsx
View File

@ -32,6 +32,8 @@ import {
isSDXLMainModelFieldInputTemplate, isSDXLMainModelFieldInputTemplate,
isSDXLRefinerModelFieldInputInstance, isSDXLRefinerModelFieldInputInstance,
isSDXLRefinerModelFieldInputTemplate, isSDXLRefinerModelFieldInputTemplate,
isSpandrelImageToImageModelFieldInputInstance,
isSpandrelImageToImageModelFieldInputTemplate,
isStringFieldInputInstance, isStringFieldInputInstance,
isStringFieldInputTemplate, isStringFieldInputTemplate,
isT2IAdapterModelFieldInputInstance, isT2IAdapterModelFieldInputInstance,
@ -54,6 +56,7 @@ import NumberFieldInputComponent from './inputs/NumberFieldInputComponent';
import RefinerModelFieldInputComponent from './inputs/RefinerModelFieldInputComponent'; import RefinerModelFieldInputComponent from './inputs/RefinerModelFieldInputComponent';
import SchedulerFieldInputComponent from './inputs/SchedulerFieldInputComponent'; import SchedulerFieldInputComponent from './inputs/SchedulerFieldInputComponent';
import SDXLMainModelFieldInputComponent from './inputs/SDXLMainModelFieldInputComponent'; import SDXLMainModelFieldInputComponent from './inputs/SDXLMainModelFieldInputComponent';
import SpandrelImageToImageModelFieldInputComponent from './inputs/SpandrelImageToImageModelFieldInputComponent';
import StringFieldInputComponent from './inputs/StringFieldInputComponent'; import StringFieldInputComponent from './inputs/StringFieldInputComponent';
import T2IAdapterModelFieldInputComponent from './inputs/T2IAdapterModelFieldInputComponent'; import T2IAdapterModelFieldInputComponent from './inputs/T2IAdapterModelFieldInputComponent';
import VAEModelFieldInputComponent from './inputs/VAEModelFieldInputComponent'; import VAEModelFieldInputComponent from './inputs/VAEModelFieldInputComponent';
@ -125,6 +128,20 @@ const InputFieldRenderer = ({ nodeId, fieldName }: InputFieldProps) => {
if (isT2IAdapterModelFieldInputInstance(fieldInstance) && isT2IAdapterModelFieldInputTemplate(fieldTemplate)) { if (isT2IAdapterModelFieldInputInstance(fieldInstance) && isT2IAdapterModelFieldInputTemplate(fieldTemplate)) {
return <T2IAdapterModelFieldInputComponent nodeId={nodeId} field={fieldInstance} fieldTemplate={fieldTemplate} />; return <T2IAdapterModelFieldInputComponent nodeId={nodeId} field={fieldInstance} fieldTemplate={fieldTemplate} />;
} }
if (
isSpandrelImageToImageModelFieldInputInstance(fieldInstance) &&
isSpandrelImageToImageModelFieldInputTemplate(fieldTemplate)
) {
return (
<SpandrelImageToImageModelFieldInputComponent
nodeId={nodeId}
field={fieldInstance}
fieldTemplate={fieldTemplate}
/>
);
}
if (isColorFieldInputInstance(fieldInstance) && isColorFieldInputTemplate(fieldTemplate)) { if (isColorFieldInputInstance(fieldInstance) && isColorFieldInputTemplate(fieldTemplate)) {
return <ColorFieldInputComponent nodeId={nodeId} field={fieldInstance} fieldTemplate={fieldTemplate} />; return <ColorFieldInputComponent nodeId={nodeId} field={fieldInstance} fieldTemplate={fieldTemplate} />;
} }

55
invokeai/frontend/web/src/features/nodes/components/flow/nodes/Invocation/fields/inputs/SpandrelImageToImageModelFieldInputComponent.tsx Normal file
View File

@ -0,0 +1,55 @@
import { Combobox, FormControl, Tooltip } from '@invoke-ai/ui-library';
import { useAppDispatch } from 'app/store/storeHooks';
import { useGroupedModelCombobox } from 'common/hooks/useGroupedModelCombobox';
import { fieldSpandrelImageToImageModelValueChanged } from 'features/nodes/store/nodesSlice';
import type {
SpandrelImageToImageModelFieldInputInstance,
SpandrelImageToImageModelFieldInputTemplate,
} from 'features/nodes/types/field';
import { memo, useCallback } from 'react';
import { useSpandrelImageToImageModels } from 'services/api/hooks/modelsByType';
import type { SpandrelImageToImageModelConfig } from 'services/api/types';
import type { FieldComponentProps } from './types';
const SpandrelImageToImageModelFieldInputComponent = (
props: FieldComponentProps<SpandrelImageToImageModelFieldInputInstance, SpandrelImageToImageModelFieldInputTemplate>
) => {
const { nodeId, field } = props;
const dispatch = useAppDispatch();
const [modelConfigs, { isLoading }] = useSpandrelImageToImageModels();
const _onChange = useCallback(
(value: SpandrelImageToImageModelConfig | null) => {
if (!value) {
return;
}
dispatch(
fieldSpandrelImageToImageModelValueChanged({
nodeId,
fieldName: field.name,
value,
})
);
},
[dispatch, field.name, nodeId]
);
const { options, value, onChange } = useGroupedModelCombobox({
modelConfigs,
onChange: _onChange,
selectedModel: field.value,
isLoading,
});
return (
<Tooltip label={value?.description}>
<FormControl className="nowheel nodrag" isInvalid={!value}>
<Combobox value={value} placeholder="Pick one" options={options} onChange={onChange} />
</FormControl>
</Tooltip>
);
};
export default memo(SpandrelImageToImageModelFieldInputComponent);

9
invokeai/frontend/web/src/features/nodes/store/nodesSlice.ts
View File

@ -19,6 +19,7 @@ import type {
ModelIdentifierFieldValue, ModelIdentifierFieldValue,
SchedulerFieldValue, SchedulerFieldValue,
SDXLRefinerModelFieldValue, SDXLRefinerModelFieldValue,
SpandrelImageToImageModelFieldValue,
StatefulFieldValue, StatefulFieldValue,
StringFieldValue, StringFieldValue,
T2IAdapterModelFieldValue, T2IAdapterModelFieldValue,
@ -39,6 +40,7 @@ import {
zModelIdentifierFieldValue, zModelIdentifierFieldValue,
zSchedulerFieldValue, zSchedulerFieldValue,
zSDXLRefinerModelFieldValue, zSDXLRefinerModelFieldValue,
zSpandrelImageToImageModelFieldValue,
zStatefulFieldValue, zStatefulFieldValue,
zStringFieldValue, zStringFieldValue,
zT2IAdapterModelFieldValue, zT2IAdapterModelFieldValue,
@ -333,6 +335,12 @@ export const nodesSlice = createSlice({
fieldT2IAdapterModelValueChanged: (state, action: FieldValueAction<T2IAdapterModelFieldValue>) => { fieldT2IAdapterModelValueChanged: (state, action: FieldValueAction<T2IAdapterModelFieldValue>) => {
fieldValueReducer(state, action, zT2IAdapterModelFieldValue); fieldValueReducer(state, action, zT2IAdapterModelFieldValue);
}, },
fieldSpandrelImageToImageModelValueChanged: (
state,
action: FieldValueAction<SpandrelImageToImageModelFieldValue>
) => {
fieldValueReducer(state, action, zSpandrelImageToImageModelFieldValue);
},
fieldEnumModelValueChanged: (state, action: FieldValueAction<EnumFieldValue>) => { fieldEnumModelValueChanged: (state, action: FieldValueAction<EnumFieldValue>) => {
fieldValueReducer(state, action, zEnumFieldValue); fieldValueReducer(state, action, zEnumFieldValue);
}, },
@ -384,6 +392,7 @@ export const {
fieldImageValueChanged, fieldImageValueChanged,
fieldIPAdapterModelValueChanged, fieldIPAdapterModelValueChanged,
fieldT2IAdapterModelValueChanged, fieldT2IAdapterModelValueChanged,
fieldSpandrelImageToImageModelValueChanged,
fieldLabelChanged, fieldLabelChanged,
fieldLoRAModelValueChanged, fieldLoRAModelValueChanged,
fieldModelIdentifierValueChanged, fieldModelIdentifierValueChanged,

1
invokeai/frontend/web/src/features/nodes/types/common.ts
View File

@ -66,6 +66,7 @@ const zModelType = z.enum([
'embedding', 'embedding',
'onnx', 'onnx',
'clip_vision', 'clip_vision',
'spandrel_image_to_image',
]); ]);
const zSubModelType = z.enum([ const zSubModelType = z.enum([
'unet', 'unet',

2
invokeai/frontend/web/src/features/nodes/types/constants.ts
View File

@ -38,6 +38,7 @@ export const MODEL_TYPES = [
'VAEField', 'VAEField',
'CLIPField', 'CLIPField',
'T2IAdapterModelField', 'T2IAdapterModelField',
'SpandrelImageToImageModelField',
]; ];
/** /**
@ -62,6 +63,7 @@ export const FIELD_COLORS: { [key: string]: string } = {
MainModelField: 'teal.500', MainModelField: 'teal.500',
SDXLMainModelField: 'teal.500', SDXLMainModelField: 'teal.500',
SDXLRefinerModelField: 'teal.500', SDXLRefinerModelField: 'teal.500',
SpandrelImageToImageModelField: 'teal.500',
StringField: 'yellow.500', StringField: 'yellow.500',
T2IAdapterField: 'teal.500', T2IAdapterField: 'teal.500',
T2IAdapterModelField: 'teal.500', T2IAdapterModelField: 'teal.500',

36
invokeai/frontend/web/src/features/nodes/types/field.ts
View File

@ -139,6 +139,10 @@ const zT2IAdapterModelFieldType = zFieldTypeBase.extend({
name: z.literal('T2IAdapterModelField'), name: z.literal('T2IAdapterModelField'),
originalType: zStatelessFieldType.optional(), originalType: zStatelessFieldType.optional(),
}); });
const zSpandrelImageToImageModelFieldType = zFieldTypeBase.extend({
name: z.literal('SpandrelImageToImageModelField'),
originalType: zStatelessFieldType.optional(),
});
const zSchedulerFieldType = zFieldTypeBase.extend({ const zSchedulerFieldType = zFieldTypeBase.extend({
name: z.literal('SchedulerField'), name: z.literal('SchedulerField'),
originalType: zStatelessFieldType.optional(), originalType: zStatelessFieldType.optional(),
@ -160,6 +164,7 @@ const zStatefulFieldType = z.union([
zControlNetModelFieldType, zControlNetModelFieldType,
zIPAdapterModelFieldType, zIPAdapterModelFieldType,
zT2IAdapterModelFieldType, zT2IAdapterModelFieldType,
zSpandrelImageToImageModelFieldType,
zColorFieldType, zColorFieldType,
zSchedulerFieldType, zSchedulerFieldType,
]); ]);
@ -581,6 +586,33 @@ export const isT2IAdapterModelFieldInputTemplate = (val: unknown): val is T2IAda
zT2IAdapterModelFieldInputTemplate.safeParse(val).success; zT2IAdapterModelFieldInputTemplate.safeParse(val).success;
// #endregion // #endregion
// #region SpandrelModelToModelField
export const zSpandrelImageToImageModelFieldValue = zModelIdentifierField.optional();
const zSpandrelImageToImageModelFieldInputInstance = zFieldInputInstanceBase.extend({
value: zSpandrelImageToImageModelFieldValue,
});
const zSpandrelImageToImageModelFieldInputTemplate = zFieldInputTemplateBase.extend({
type: zSpandrelImageToImageModelFieldType,
originalType: zFieldType.optional(),
default: zSpandrelImageToImageModelFieldValue,
});
const zSpandrelImageToImageModelFieldOutputTemplate = zFieldOutputTemplateBase.extend({
type: zSpandrelImageToImageModelFieldType,
});
export type SpandrelImageToImageModelFieldValue = z.infer<typeof zSpandrelImageToImageModelFieldValue>;
export type SpandrelImageToImageModelFieldInputInstance = z.infer<typeof zSpandrelImageToImageModelFieldInputInstance>;
export type SpandrelImageToImageModelFieldInputTemplate = z.infer<typeof zSpandrelImageToImageModelFieldInputTemplate>;
export const isSpandrelImageToImageModelFieldInputInstance = (
val: unknown
): val is SpandrelImageToImageModelFieldInputInstance =>
zSpandrelImageToImageModelFieldInputInstance.safeParse(val).success;
export const isSpandrelImageToImageModelFieldInputTemplate = (
val: unknown
): val is SpandrelImageToImageModelFieldInputTemplate =>
zSpandrelImageToImageModelFieldInputTemplate.safeParse(val).success;
// #endregion
// #region SchedulerField // #region SchedulerField
export const zSchedulerFieldValue = zSchedulerField.optional(); export const zSchedulerFieldValue = zSchedulerField.optional();
@ -667,6 +699,7 @@ export const zStatefulFieldValue = z.union([
zControlNetModelFieldValue, zControlNetModelFieldValue,
zIPAdapterModelFieldValue, zIPAdapterModelFieldValue,
zT2IAdapterModelFieldValue, zT2IAdapterModelFieldValue,
zSpandrelImageToImageModelFieldValue,
zColorFieldValue, zColorFieldValue,
zSchedulerFieldValue, zSchedulerFieldValue,
]); ]);
@ -694,6 +727,7 @@ const zStatefulFieldInputInstance = z.union([
zControlNetModelFieldInputInstance, zControlNetModelFieldInputInstance,
zIPAdapterModelFieldInputInstance, zIPAdapterModelFieldInputInstance,
zT2IAdapterModelFieldInputInstance, zT2IAdapterModelFieldInputInstance,
zSpandrelImageToImageModelFieldInputInstance,
zColorFieldInputInstance, zColorFieldInputInstance,
zSchedulerFieldInputInstance, zSchedulerFieldInputInstance,
]); ]);
@ -722,6 +756,7 @@ const zStatefulFieldInputTemplate = z.union([
zControlNetModelFieldInputTemplate, zControlNetModelFieldInputTemplate,
zIPAdapterModelFieldInputTemplate, zIPAdapterModelFieldInputTemplate,
zT2IAdapterModelFieldInputTemplate, zT2IAdapterModelFieldInputTemplate,
zSpandrelImageToImageModelFieldInputTemplate,
zColorFieldInputTemplate, zColorFieldInputTemplate,
zSchedulerFieldInputTemplate, zSchedulerFieldInputTemplate,
zStatelessFieldInputTemplate, zStatelessFieldInputTemplate,
@ -751,6 +786,7 @@ const zStatefulFieldOutputTemplate = z.union([
zControlNetModelFieldOutputTemplate, zControlNetModelFieldOutputTemplate,
zIPAdapterModelFieldOutputTemplate, zIPAdapterModelFieldOutputTemplate,
zT2IAdapterModelFieldOutputTemplate, zT2IAdapterModelFieldOutputTemplate,
zSpandrelImageToImageModelFieldOutputTemplate,
zColorFieldOutputTemplate, zColorFieldOutputTemplate,
zSchedulerFieldOutputTemplate, zSchedulerFieldOutputTemplate,
]); ]);

1
invokeai/frontend/web/src/features/nodes/util/schema/buildFieldInputInstance.ts
View File

@ -18,6 +18,7 @@ const FIELD_VALUE_FALLBACK_MAP: Record<StatefulFieldType['name'], FieldValue> =
SDXLRefinerModelField: undefined, SDXLRefinerModelField: undefined,
StringField: '', StringField: '',
T2IAdapterModelField: undefined, T2IAdapterModelField: undefined,
SpandrelImageToImageModelField: undefined,
VAEModelField: undefined, VAEModelField: undefined,
ControlNetModelField: undefined, ControlNetModelField: undefined,
}; };

13
invokeai/frontend/web/src/features/nodes/util/schema/buildFieldInputTemplate.ts
View File

@ -17,6 +17,7 @@ import type {
SchedulerFieldInputTemplate, SchedulerFieldInputTemplate,
SDXLMainModelFieldInputTemplate, SDXLMainModelFieldInputTemplate,
SDXLRefinerModelFieldInputTemplate, SDXLRefinerModelFieldInputTemplate,
SpandrelImageToImageModelFieldInputTemplate,
StatefulFieldType, StatefulFieldType,
StatelessFieldInputTemplate, StatelessFieldInputTemplate,
StringFieldInputTemplate, StringFieldInputTemplate,
@ -263,6 +264,17 @@ const buildT2IAdapterModelFieldInputTemplate: FieldInputTemplateBuilder<T2IAdapt
return template; return template;
}; };
const buildSpandrelImageToImageModelFieldInputTemplate: FieldInputTemplateBuilder<
SpandrelImageToImageModelFieldInputTemplate
> = ({ schemaObject, baseField, fieldType }) => {
const template: SpandrelImageToImageModelFieldInputTemplate = {
...baseField,
type: fieldType,
default: schemaObject.default ?? undefined,
};
return template;
};
const buildBoardFieldInputTemplate: FieldInputTemplateBuilder<BoardFieldInputTemplate> = ({ const buildBoardFieldInputTemplate: FieldInputTemplateBuilder<BoardFieldInputTemplate> = ({
schemaObject, schemaObject,
baseField, baseField,
@ -377,6 +389,7 @@ export const TEMPLATE_BUILDER_MAP: Record<StatefulFieldType['name'], FieldInputT
SDXLRefinerModelField: buildRefinerModelFieldInputTemplate, SDXLRefinerModelField: buildRefinerModelFieldInputTemplate,
StringField: buildStringFieldInputTemplate, StringField: buildStringFieldInputTemplate,
T2IAdapterModelField: buildT2IAdapterModelFieldInputTemplate, T2IAdapterModelField: buildT2IAdapterModelFieldInputTemplate,
SpandrelImageToImageModelField: buildSpandrelImageToImageModelFieldInputTemplate,
VAEModelField: buildVAEModelFieldInputTemplate, VAEModelField: buildVAEModelFieldInputTemplate,
} as const; } as const;

1
invokeai/frontend/web/src/features/nodes/util/workflow/validateWorkflow.ts
View File

@ -35,6 +35,7 @@ const MODEL_FIELD_TYPES = [
'ControlNetModelField', 'ControlNetModelField',
'IPAdapterModelField', 'IPAdapterModelField',
'T2IAdapterModelField', 'T2IAdapterModelField',
'SpandrelImageToImageModelField',
]; ];
/** /**

2
invokeai/frontend/web/src/services/api/hooks/modelsByType.ts
View File

@ -11,6 +11,7 @@ import {
isNonSDXLMainModelConfig, isNonSDXLMainModelConfig,
isRefinerMainModelModelConfig, isRefinerMainModelModelConfig,
isSDXLMainModelModelConfig, isSDXLMainModelModelConfig,
isSpandrelImageToImageModelConfig,
isT2IAdapterModelConfig, isT2IAdapterModelConfig,
isTIModelConfig, isTIModelConfig,
isVAEModelConfig, isVAEModelConfig,
@ -39,6 +40,7 @@ export const useLoRAModels = buildModelsHook(isLoRAModelConfig);
export const useControlNetAndT2IAdapterModels = buildModelsHook(isControlNetOrT2IAdapterModelConfig); export const useControlNetAndT2IAdapterModels = buildModelsHook(isControlNetOrT2IAdapterModelConfig);
export const useControlNetModels = buildModelsHook(isControlNetModelConfig); export const useControlNetModels = buildModelsHook(isControlNetModelConfig);
export const useT2IAdapterModels = buildModelsHook(isT2IAdapterModelConfig); export const useT2IAdapterModels = buildModelsHook(isT2IAdapterModelConfig);
export const useSpandrelImageToImageModels = buildModelsHook(isSpandrelImageToImageModelConfig);
export const useIPAdapterModels = buildModelsHook(isIPAdapterModelConfig); export const useIPAdapterModels = buildModelsHook(isIPAdapterModelConfig);
export const useEmbeddingModels = buildModelsHook(isTIModelConfig); export const useEmbeddingModels = buildModelsHook(isTIModelConfig);
export const useVAEModels = buildModelsHook(isVAEModelConfig); export const useVAEModels = buildModelsHook(isVAEModelConfig);

410
invokeai/frontend/web/src/services/api/schema.ts
View File

File diff suppressed because one or more lines are too long

8
invokeai/frontend/web/src/services/api/types.ts
View File

@ -51,6 +51,7 @@ export type VAEModelConfig = S['VAECheckpointConfig'] | S['VAEDiffusersConfig'];
export type ControlNetModelConfig = S['ControlNetDiffusersConfig'] | S['ControlNetCheckpointConfig']; export type ControlNetModelConfig = S['ControlNetDiffusersConfig'] | S['ControlNetCheckpointConfig'];
export type IPAdapterModelConfig = S['IPAdapterInvokeAIConfig'] | S['IPAdapterCheckpointConfig']; export type IPAdapterModelConfig = S['IPAdapterInvokeAIConfig'] | S['IPAdapterCheckpointConfig'];
export type T2IAdapterModelConfig = S['T2IAdapterConfig']; export type T2IAdapterModelConfig = S['T2IAdapterConfig'];
export type SpandrelImageToImageModelConfig = S['SpandrelImageToImageConfig'];
type TextualInversionModelConfig = S['TextualInversionFileConfig'] | S['TextualInversionFolderConfig']; type TextualInversionModelConfig = S['TextualInversionFileConfig'] | S['TextualInversionFolderConfig'];
type DiffusersModelConfig = S['MainDiffusersConfig']; type DiffusersModelConfig = S['MainDiffusersConfig'];
type CheckpointModelConfig = S['MainCheckpointConfig']; type CheckpointModelConfig = S['MainCheckpointConfig'];
@ -62,6 +63,7 @@ export type AnyModelConfig =
| ControlNetModelConfig | ControlNetModelConfig
| IPAdapterModelConfig | IPAdapterModelConfig
| T2IAdapterModelConfig | T2IAdapterModelConfig
| SpandrelImageToImageModelConfig
| TextualInversionModelConfig | TextualInversionModelConfig
| MainModelConfig | MainModelConfig
| CLIPVisionDiffusersConfig; | CLIPVisionDiffusersConfig;
@ -86,6 +88,12 @@ export const isT2IAdapterModelConfig = (config: AnyModelConfig): config is T2IAd
return config.type === 't2i_adapter'; return config.type === 't2i_adapter';
}; };
export const isSpandrelImageToImageModelConfig = (
config: AnyModelConfig
): config is SpandrelImageToImageModelConfig => {
return config.type === 'spandrel_image_to_image';
};
export const isControlAdapterModelConfig = ( export const isControlAdapterModelConfig = (
config: AnyModelConfig config: AnyModelConfig
): config is ControlNetModelConfig | T2IAdapterModelConfig | IPAdapterModelConfig => { ): config is ControlNetModelConfig | T2IAdapterModelConfig | IPAdapterModelConfig => {

2
invokeai/version/invokeai_version.py
View File

@ -1 +1 @@
__version__ = "4.2.6a1" __version__ = "4.2.6post1"

1
pyproject.toml
View File

@ -46,6 +46,7 @@ dependencies = [
"opencv-python==4.9.0.80", "opencv-python==4.9.0.80",
"pytorch-lightning==2.1.3", "pytorch-lightning==2.1.3",
"safetensors==0.4.3", "safetensors==0.4.3",
"spandrel==0.3.4",
"timm==0.6.13", # needed to override timm latest in controlnet_aux, see https://github.com/isl-org/ZoeDepth/issues/26 "timm==0.6.13", # needed to override timm latest in controlnet_aux, see https://github.com/isl-org/ZoeDepth/issues/26
"torch==2.2.2", "torch==2.2.2",
"torchmetrics==0.11.4", "torchmetrics==0.11.4",