678f933d38
maximum_batch_area for xformers.
...
Remove useless code.
2023-05-06 19:28:46 -04:00
908dc1d5a8
Add a total_steps value to sampler callback.
2023-05-03 12:58:10 -04:00
93c64afaa9
Use sampler callback instead of tqdm hook for progress bar.
2023-05-02 23:00:49 -04:00
ba8a4c3667
Change latent resolution step to 8.
2023-05-02 14:17:51 -04:00
35f636b6c7
Expose grow_mask_by in VAEEncodeForInpaint.
...
The mask is dilated by grow_mask_by pixels after being applied to the pixel
space image. This helps reduce seams caused by inpainting. Higher value
means less seams.
2023-05-02 00:53:15 -04:00
0aa667ed33
Fix ConditioningAverage.
2023-04-30 17:33:43 -04:00
29c8f1a344
Conditioning Average ( #495 )
...
* first commit
* fixed a bunch of things missing in initial commit.
* parameters renamed for clarity
* renamed node, attempted update cond list
* to_strength removed, it is now normalized
* removed comments and prints. Attempted to apply to every cond in list again but no luck
* fixed repeating frames after batch using deepcopy
* Revert "fixed repeating frames after batch using deepcopy"
This reverts commit 1086d6a0e1
2023-04-30 17:33:15 -04:00
c66db06763
Make ConditioningSetMask area option a bit more clear.
...
Make ConditioningSetArea override the set_area_to_bounds.
2023-04-29 20:25:05 -04:00
071011aebe
Mask strength should be separate from area strength.
2023-04-29 20:06:53 -04:00
af02393c2a
Default to sampling entire image
...
By default, when applying a mask to a condition, the entire image will
still be used for sampling. The new "set_area_to_bounds" option on the
node will allow the user to automatically limit conditioning to the
bounds of the mask.
I've also removed the dependency on torchvision for calculating bounding
boxes. I've taken the opportunity to fix some frustrating details in the
other version:
1. An all-0 mask will no longer cause an error
2. Indices are returned as integers instead of floats so they can be
used to index into tensors.
2023-04-29 00:16:58 -07:00
e214c917ae
Add Condition by Mask node
...
This PR adds support for a Condition by Mask node. This node allows
conditioning to be limited to a non-rectangle area.
2023-04-27 20:03:27 -07:00
aa57136dae
Some fixes to the batch masks PR.
2023-04-25 01:12:40 -04:00
c7c1f0d074
Merge branch 'multiple_inpaint_masks' of https://github.com/guill/ComfyUI
...
# Conflicts:
# nodes.py
2023-04-25 00:54:15 -04:00
c50208a703
Refactor more code to sample.py
2023-04-24 23:25:51 -04:00
a2236a0c2a
Merge branch 'master' of https://github.com/BlenderNeko/ComfyUI
2023-04-24 17:57:27 -04:00
d9b1595f85
made sample functions more explicit
2023-04-24 12:53:10 +02:00
ccad603b2e
Add a way for nodes to validate their own inputs.
2023-04-23 16:03:26 -04:00
f7a8218814
Add clipspace feature. ( #541 )
...
* Add clipspace feature.
* feat: copy content to clipspace
* feat: paste content from clipspace
Extend validation to allow for validating annotated_path in addition to other parameters.
Add support for annotated_filepath in folder_paths function.
Generalize the '/upload/image' API to allow for uploading images to the 'input', 'temp', or 'output' directories.
* rename contentClipboard -> clipspace
* Do deep copy for imgs on copy to clipspace.
* add original_imgs into clipspace
* Preserve the original image when 'imgs' are modified
* robust patch & refactoring folder_paths about annotated_filepath
* Only show the Paste menu if the ComfyApp.clipspace is not empty
* instant refresh on paste
force triggering 'changed' on paste action
* subfolder fix on paste logic
attach subfolder if subfolder isn't empty
---------
Co-authored-by: Lt.Dr.Data <lt.dr.data@gmail.com>
2023-04-23 15:58:55 -04:00
5818539743
add docstrings
2023-04-23 20:09:09 +02:00
8d2de420d3
Merge branch 'master' of https://github.com/BlenderNeko/ComfyUI
2023-04-23 20:02:18 +02:00
2a09e2aa27
refactor/split various bits of code for sampling
2023-04-23 20:02:08 +02:00
5282f56434
Implement Linear hypernetworks.
...
Add a HypernetworkLoader node to use hypernetworks.
2023-04-23 12:35:25 -04:00
ee030d281b
Add support for multiple unique inpainting masks
...
This enables workflows like "Inpaint at full resolution" when using
batch sizes greater than 1.
2023-04-22 16:02:26 -07:00
31e60adb28
Add GLIGEN example to README.
2023-04-20 17:30:10 -04:00
3696d1699a
Add support for GLIGEN textbox model.
2023-04-19 11:06:32 -04:00
f03dade5ab
Fix bug.
2023-04-17 18:19:57 -04:00
6f7852bc47
Add a LatentFromBatch node to pick a single latent from a batch.
...
Works before and after sampling.
2023-04-17 17:28:07 -04:00
74fc7b7726
custom_nodes paths can now be set in the extra_model_paths.yaml
2023-04-16 01:46:43 -04:00
73c3e11e83
Fix model_management import so it doesn't get executed twice.
2023-04-15 19:04:33 -04:00
f48f0872e2
Refactor: move nodes_mask_convertion nodes to nodes_mask.
2023-04-14 00:21:01 -04:00
e1db7a2038
Merge branch 'image-to-mask' of https://github.com/missionfloyd/ComfyUI
...
# Conflicts:
# nodes.py
2023-04-14 00:15:48 -04:00
1a7cda715b
Revert LatentComposite.
2023-04-14 00:14:35 -04:00
fed4a70b8e
Merge branch 'master' of https://github.com/mligaintart/ComfyUI
2023-04-14 00:12:58 -04:00
d2337a86fe
remove extra semi-colon
2023-04-13 16:38:02 -04:00
9371924e65
Move mask conversion to separate file
2023-04-13 03:11:17 -06:00
45b907fbf1
Merge branch 'comfyanonymous:master' into image-to-mask
2023-04-13 03:07:08 -06:00
f15b4a7ffa
Merge branch 'fix-LoadImageMask' of https://github.com/flyingshutter/ComfyUI
2023-04-12 01:24:53 -04:00
e1d289c1ec
use slice instead of torch.select()
2023-04-11 20:26:24 -06:00
e12fb88b1b
Image/mask conversion nodes
2023-04-11 16:49:39 -06:00
24d53992c6
Rename.
2023-04-08 15:53:01 -04:00
60ac983238
Merge branch 'master' of https://github.com/city96/ComfyUI
2023-04-08 15:50:26 -04:00
44fea05064
Cleanup.
2023-04-07 02:31:46 -04:00
58ed0f2da4
Fix loading SD1.5 diffusers checkpoint.
2023-04-07 01:30:33 -04:00
7734d65f22
fix loading alt folders
2023-04-06 22:02:26 -07:00
f51b7a92c7
search all diffusers paths (oops)
2023-04-06 21:48:58 -07:00
72a8973bd5
allow configurable path for diffusers models
2023-04-06 21:45:08 -07:00
022a9f271b
Adds masking to Latent Composite, and provides new masking utilities to
...
allow better compositing.
2023-04-06 15:18:20 -04:00
9d1c610b89
make LoadImagesMask work with non RGBA images
2023-04-06 19:02:28 +02:00
42fd67b5cb
use precision determined by model management
2023-04-06 00:28:06 -07:00
3d16077e38
empty list if diffusers directory doesn't exist
2023-04-06 00:24:52 -07:00
60127a8304
diffusers loader
2023-04-05 23:57:31 -07:00
f816964847
Add a way to set output directory with --output-directory
2023-04-05 14:29:30 -04:00
10ad4c1d17
Move unclip stuff out of _for_testing
2023-04-04 22:48:11 -04:00
cadef9ff61
Merge branch 'nodes-post-processing' of https://github.com/EllangoK/ComfyUI
2023-04-04 13:51:26 -04:00
f50b1fec69
Add noise augmentation setting to unCLIPConditioning.
2023-04-03 13:50:29 -04:00
4c7a9dbcb6
adds Blend, Blur, Dither, Sharpen nodes
2023-04-02 18:44:27 -04:00
6be7c64245
Merge branch 'custom_routes' of https://github.com/pythongosssss/ComfyUI
2023-04-02 01:55:05 -04:00
809bcc8ceb
Add support for unCLIP SD2.x models.
...
See _for_testing/unclip in the UI for the new nodes.
unCLIPCheckpointLoader is used to load them.
unCLIPConditioning is used to add the image cond and takes as input a
CLIPVisionEncode output which has been moved to the conditioning section.
2023-04-01 23:19:15 -04:00
38833ceb62
Merge remote-tracking branch 'origin/master' into custom_routes
2023-04-01 12:44:40 +01:00
313f1f83a6
Tweak server/custom node load order
...
- Load custom nodes after creating server
- Add routes after loading custom nodes
Custom nodes can now add routes via PromptServer.instance
2023-04-01 12:44:29 +01:00
18a6c1db33
Add a TomePatchModel node to the _for_testing section.
...
Tome increases sampling speed at the expense of quality.
2023-03-31 17:19:58 -04:00
9ac95e6ac7
Add human-readable name support for custom nodes
2023-03-31 07:05:17 +02:00
5bcbc007dd
Add human-readable names for nodes
2023-03-30 23:13:58 +02:00
bb1503ed63
Bugfix/include optional node inputs ( #271 )
...
* Minor changes and extra nodes
* Added Preview Image for Nodes
* Add a delete images button on main floating menu
* Add a confirmation dialog
* Remove DeleteAll, Remove Custom KSampler, Remove Image List
* Remove Image and Custom Json Load
* Remove Custom Nodes
* remove patch.diff
2023-03-27 01:16:22 -04:00
48efadeccf
Style.
2023-03-26 15:17:53 -04:00
8a35c12e8c
compute %width% and %height% in filepath when saving
2023-03-26 13:10:20 +02:00
9287770c66
Allow negative Lora strengths.
2023-03-25 21:31:39 -04:00
e651be551c
Merge branch 'allow-symbolic-link' of https://github.com/jn-jairo/ComfyUI
2023-03-25 17:30:22 -04:00
a30526aec1
Set default value of feathering to 40.
2023-03-25 04:49:58 -04:00
3b87302958
nodes: ImagePadForOutpaint: clean imports
2023-03-25 16:32:48 +08:00
4c01386c19
nodes: ImagePadForOutpaint: fix feathering, ignore edges not expanded
2023-03-25 16:27:47 +08:00
4580f3e644
nodes: add feathering to to ImagePadForOutpaint
2023-03-24 22:39:33 +08:00
4f24e6a13b
Allow output folder to be a symbolic link
2023-03-23 21:25:21 -03:00
3ebf7452c3
nodes: add ImagePadForOutpaint
2023-03-23 23:33:35 +08:00
cc127eeabd
Lower PNG compression for more speed.
...
After some quick tests compress_level=4 seems to give the best compression
performance ratio for stable diffusion images.
It's 10x faster than max compression for an image size increase of only
2.5%.
2023-03-23 00:40:48 -04:00
aae9fe0cf9
Increase max res to 8192x8192 since 4096x4096 wasn't enough for some.
2023-03-22 12:22:48 -04:00
aa2ddfabb9
Fix bug with CLIPLoader.
2023-03-21 03:11:18 -04:00
e85fcb822b
Small fix.
2023-03-20 14:55:28 -04:00
0abe001edd
Merge branch 'save-images' of https://github.com/m957ymj75urz/ComfyUI
2023-03-20 14:29:45 -04:00
d1138e8ba0
rework payload from server
2023-03-19 12:54:29 +01:00
50099bcd96
Support multiple paths for embeddings.
2023-03-18 03:08:43 -04:00
2e73367f45
Merge T2IAdapterLoader and ControlNetLoader.
...
Workflows will be auto updated.
2023-03-17 18:17:59 -04:00
e1a9e26968
Add folder_paths so models can be in multiple paths.
2023-03-17 18:01:11 -04:00
bf1dc1d9bc
Fix VAEEncodeForInpaint modifying source image.
2023-03-16 17:10:08 -04:00
16001e9429
fixes for specific paths
2023-03-16 19:48:59 +01:00
d834eed7d0
Fix bug when applying controlnet to negative prompt.
2023-03-15 17:58:13 -04:00
629272c0ca
resolve conflict with the new PreviewImage node
2023-03-15 12:10:45 +01:00
3d2f60b315
Merge branch 'master' into save-images
2023-03-15 10:48:15 +00:00
6d44cf74e3
Make it more clear the recommended way to update the standalone build.
2023-03-14 19:14:18 -04:00
8537ab6f45
tidy
2023-03-14 19:42:28 +00:00
6a6256a75c
Changed flag to new node
2023-03-14 19:28:07 +00:00
6daf9bb22c
switch to realpath to check path traversal
2023-03-14 09:27:53 +01:00
5b425aaa40
remove subfolder widget and use filename_prefix
2023-03-14 09:08:54 +01:00
6db777b348
Added ability to save images to temp dir
2023-03-13 19:34:05 +00:00
70f9bd1408
fix makedirs on save
2023-03-12 20:23:46 +01:00
195d7aec9f
can specify a subfolder in the SaveImage node
2023-03-12 19:51:39 +01:00
c8f1acc4eb
Put image upscaling nodes in image/upscaling category.
2023-03-11 18:10:36 -05:00
e33dc2b33b
Add a VAEEncodeTiled node.
2023-03-11 15:28:15 -05:00
905857edd8
Take some code from chainner to implement ESRGAN and other upscale models.
2023-03-11 13:09:28 -05:00
8c4ccb55d1
Small refactor of custom node loading code.
2023-03-11 12:49:41 -05:00
1f717903bc
Prevent potential int overflow in SaveImage
2023-03-11 12:48:28 -05:00
99eb777f5d
Make the LoadImage node output the alpha as a mask.
2023-03-09 14:07:55 -05:00
fa07426569
Remove random spaces
2023-03-09 18:18:08 +00:00
99abc0eb2e
Changed to upload to input dir
...
Fixed jpg
Added dupe support
Changed to use existing nodes
2023-03-09 17:57:59 +00:00
4a326a2548
Added UploadImage+Mask nodes
2023-03-08 22:07:44 +00:00
0a0c769d6d
Try to fix import issue.
2023-03-06 14:41:42 -05:00
09db31309a
Merge remote-tracking branch 'origin/master' into frontendrefactor
2023-03-06 17:02:34 +00:00
afff30fc0a
Add --cpu to use the cpu for inference.
2023-03-06 10:50:50 -05:00
8515d96328
It's more logical for the StyleModelApply to append it directly.
2023-03-06 01:48:18 -05:00
7ec1dd25a2
A tiny bit of reorganizing.
2023-03-06 01:30:17 -05:00
47acb3d73e
Implement support for t2i style model.
...
It needs the CLIPVision model so I added CLIPVisionLoader and CLIPVisionEncode.
Put the clip vision model in models/clip_vision
Put the t2i style model in models/style_models
StyleModelLoader to load it, StyleModelApply to apply it
ConditioningAppend to append the conditioning it outputs to a positive one.
2023-03-05 18:39:25 -05:00
bba14245cb
Merge remote-tracking branch 'origin' into frontendrefactor
2023-03-03 19:30:24 +00:00
b9e3f52240
stop_at_clip_layer can be set in CLIPSetLastLayer so it can be removed
...
from the CLIPLoader node.
2023-03-03 14:27:55 -05:00
4ef4cf913f
Adding built in extensions + example
2023-03-03 18:28:34 +00:00
8586515886
Move the CheckpointLoaderSimple to the right section.
2023-03-03 13:09:44 -05:00
4215206281
Add a node to set CLIP skip.
...
Use a more simple way to detect if the model is -v prediction.
2023-03-03 13:04:36 -05:00
fed315a76a
To be really simple CheckpointLoaderSimple should pick the right type.
2023-03-03 11:07:10 -05:00
94bb0375b0
New CheckpointLoaderSimple to load checkpoints without a config.
2023-03-03 03:37:35 -05:00
c8ce599a8f
Add a button to interrupt processing to the ui.
2023-03-02 15:24:51 -05:00
69cc75fbf8
Add a way to interrupt current processing in the backend.
2023-03-02 14:42:03 -05:00
d826a0971c
Outpainting actually works now.
2023-02-27 12:02:23 -05:00
a9c57849b7
Merge branch 'comfyanonymous:master' into socketrework
2023-02-25 12:00:22 +00:00
4e6b83a80a
Add a T2IAdapterLoader node to load T2I-Adapter models.
...
They are loaded as CONTROL_NET objects because they are similar.
2023-02-25 01:24:56 -05:00
87b00b37f6
Added an experimental VAEDecodeTiled.
...
This decodes the image with the VAE in tiles which should be faster and
use less vram.
It's in the _for_testing section so I might change/remove it or even
add the functionality to the regular VAEDecode node depending on how
well it performs which means don't depend too much on it.
2023-02-24 02:10:10 -05:00
9bd7bfa648
Added workflow history
...
Moved socket output updates to all node executions
Made image rendering on nodes more generic
2023-02-23 20:12:57 +00:00
2816eb236d
Merge branch 'comfyanonymous:master' into socketrework
2023-02-23 20:07:30 +00:00
62df8dd62a
Add a node to load diff controlnets.
2023-02-22 23:22:03 -05:00
3ae61a2bca
Add a node to invert the colours of images.
...
Might be useful for controlnets.
2023-02-22 21:57:56 -05:00
a52aa9f4b5
Moved api out to server
...
Reworked sockets to use socketio
Added progress to nodes
Added highlight to active node
Added preview to saveimage node
2023-02-21 19:29:49 +00:00
d80af7ca30
ControlNetApply now stacks.
...
It can be used to apply multiple control nets at the same time.
2023-02-21 01:18:53 -05:00
8683ea4248
Merge branch 'master' into m957ymj75urz-dynamic-prompting
2023-02-20 23:49:55 -05:00
a3d0b56c61
Fix for relative imports in custom nodes.
2023-02-19 12:00:05 -05:00
e2e5c4c5b0
Add basic dynamic prompting with seeding on CLIPTextEncode node
2023-02-17 22:37:03 +01:00
71354c7c57
Sort images on LoadImageMask node.
2023-02-17 13:17:09 -05:00
1688f5024d
Fix a few issues with the custom_nodes PR.
...
There only needs to be one example in the folder.
2023-02-17 11:19:49 -05:00
fa66ece26b
Merge branch 'comfyanonymous:master' into master
2023-02-17 18:02:52 +07:00
1c5fe809c2
.
2023-02-17 17:59:16 +07:00
638f924546
Sort the list of input images.
2023-02-16 21:01:46 -05:00
6135a21ee8
Add a way to control controlnet strength.
2023-02-16 18:08:01 -05:00
4efa67fa12
Add ControlNet support.
2023-02-16 10:38:08 -05:00
bc69fb5245
Use inpaint models the proper way by using VAEEncodeForInpaint.
2023-02-15 20:44:51 -05:00
e87a8669b6
Add a LoadImageMask node to load one colour channel in an image as a mask.
2023-02-15 17:39:42 -05:00
d75003001a
Add a SetLatentNoiseMask node.
...
LATENT is now a dict that can contain properties.
2023-02-15 16:58:55 -05:00
e1a047971b
Edit comment
2023-02-15 21:50:18 +07:00
7da0147d8a
Done custom_node import error fixing
2023-02-15 21:48:10 +07:00
b67b6920d1
Fix bug
2023-02-15 21:15:54 +07:00
493671574e
Adding some suggests from the author
2023-02-14 22:20:30 +07:00
721f8510af
Change the words a bit
2023-02-13 18:24:54 +07:00
f3c6d7adf2
.
2023-02-13 18:19:42 +07:00
5ce4b9eca3
Add custom node feature
2023-02-13 18:17:40 +07:00
05ad64d22c
Add a feather option to the latent composite node.
2023-02-12 13:01:52 -05:00
642516a3a6
create output dir if none is present
2023-02-09 12:49:31 -05:00
1d9ec62cfb
Use absolute output directory path.
2023-02-09 09:59:43 -05:00
e58887dfa7
forgot windows does double backslashes for paths due to its use as escape char.
2023-02-09 01:30:06 +01:00
81082045c2
add recursive_search, swap relevant os.listdirs
2023-02-09 01:22:33 +01:00
3e22815a9a
Fix k_diffusion not getting imported from the folder.
2023-02-08 16:29:22 -05:00
a84cd0d1ad
Don't unload/reload model from CPU uselessly.
2023-02-08 03:40:43 -05:00
1de5aa6a59
Add a CLIPLoader node to load standalone clip weights.
...
Put them in models/clip
2023-02-05 15:20:18 -05:00
4225d1cb9f
Add a basic ImageScale node.
...
It's pretty much the same as the LatentUpscale node for now but for images
in pixel space.
2023-02-04 16:01:01 -05:00
bff0e11941
Add a LatentCrop node.
2023-02-04 15:21:46 -05:00
149a4de3f2
Fix potential issue if exception happens when patching model.
2023-02-03 03:55:50 -05:00
ef90e9c376
Add a LoraLoader node to apply loras to models and clip.
...
The models are modified in place before being used and unpatched after.
I think this is better than monkeypatching since it might make it easier
to use faster non pytorch unet inference in the future.
2023-02-03 02:46:24 -05:00
fef41d0a72
Add LatentComposite node.
...
This can be used to "paste" one latent image on top of the other.
2023-01-31 03:35:03 -05:00
3fa009f4cc
Add a LatentFlip node.
2023-01-31 03:28:38 -05:00
69df7eba94
Add KSamplerAdvanced node.
...
This node exposes more sampling options and makes it possible for example
to sample the first few steps on the latent image, do some operations on it
and then do the rest of the sampling steps. This can be achieved using the
start_at_step and end_at_step options.
2023-01-31 03:09:38 -05:00
f8f165e2c3
Add a LatentRotate node.
2023-01-31 02:28:07 -05:00
f73e57d881
Add support for textual inversion embedding for SD1.x CLIP.
2023-01-29 18:46:44 -05:00
2706c0b7a5
Some VAEs come in .pt files.
2023-01-28 12:28:29 -05:00
21eb33f233
Right category for latent upscale node.
2023-01-27 14:11:57 -05:00
2a87f6630d
Some categories for the nodes.
2023-01-26 12:23:15 -05:00
c4b02059d0
Add ConditioningSetArea node.
...
to apply conditioning/prompts only to a specific area of the image.
Add ConditioningCombine node.
so that multiple conditioning/prompts can be applied to the image at the
same time
2023-01-26 12:06:48 -05:00
463d0d0828
Add a center crop option to latent upscale node.
2023-01-24 17:28:44 -05:00
c1eac7bab2
Add filename_prefix to SaveImage node.
2023-01-24 03:05:04 -05:00
a34cde08cb
Nodes now support single line string properties.
...
Resize nodes automatically if they are too small because of added properties.
2023-01-24 02:10:12 -05:00
9baa48cb33
Add IS_CHANGED method to nodes to check if nodes should be executed again.
...
LoadImage.IS_CHANGED returns the hash of the image so it will execute again
if the image changed on the disk.
2023-01-22 21:46:16 -05:00
15f8da2849
Add a LoadImage node to load images for img2img.
2023-01-22 15:07:18 -05:00
220afe3310
Initial commit.
2023-01-16 22:37:14 -05:00
comfyanonymous
FizzleDorf
Jacob Segal
BlenderNeko
ltdrdata
Gavroche CryptoRUSH
missionfloyd
sALTaccount
mligaintart
flyingshutter
EllangoK
pythongosssss
City
Silversith
m957ymj75urz
Guo Y.K
Jairo Correa
m957ymj75urz
Fannovel16
BazettFraga