810 Commits

Author SHA1 Message Date
kijai
9954d6b599 Update model_optimization_nodes.py 2025-10-06 21:37:56 +03:00
kijai
08164edab3 Fix sage3 for Wan 2025-10-06 21:23:47 +03:00
kijai
b1fffd33ba bump version 2025-10-06 15:11:11 +03:00
kijai
3aff68488f Merge branch 'pr/402' 2025-10-06 15:07:36 +03:00
kijai
ead0cce23b Some fixes/adjustments 2025-10-06 15:07:08 +03:00
kijai
b64d66404d Fix some tooltips and silence GGUF loader node definition error
In case of no ComfyUI-GGUF or WanVideoWrapper is installed, there was unnecessary error printed
2025-10-06 14:49:27 +03:00
kijai
625151d43f Update mask_nodes.py 2025-10-03 19:35:15 +03:00
kijai
fb253119ed Update mask_nodes.py 2025-10-03 19:23:27 +03:00
kijai
4d411c4593 Update mask_nodes.py 2025-10-03 19:14:27 +03:00
kijai
d1f6cf77b0 optimize draw mask on image 2025-10-03 19:06:09 +03:00
kijai
b9abf5df31 optimize blockify mask node 2025-10-03 19:01:46 +03:00
kijai
bb205d809b Update lora_nodes.py 2025-10-02 01:31:04 +03:00
kijai
ab89f1c6c9 Update model_optimization_nodes.py 2025-10-02 01:18:39 +03:00
kijai
7ce88200c7 sageattn3 2025-10-02 01:09:59 +03:00
stephennilor
629a2bf423 revert accidental change to LoadVideosFromFolder node 2025-10-01 18:41:29 +08:00
stephennilor
051c10382b add optional batching of image resizing 2025-10-01 18:12:17 +08:00
kijai
9d7af919b9 Use kornia for GPU accelerated mask dilation 2025-09-23 20:15:13 +03:00
Jukka Seppänen
eeb797736b
Merge pull request #397 from stuttlepress/fix-loadvideosfromfolder-windows-import
Fix LoadVideosFromFolder VideoHelperSuite import in Windows
2025-09-23 13:37:53 +03:00
stuttlepress
11a482d3c3 Fix LoadVideosFromFolder VHS import in Windows
VideoHelperSuite import fails on Windows even when VHS is properly
installed, due to custom node module names being stored as full file
paths for some reason.
Add fallback to sys.modules search when standard imports fail.
2025-09-23 03:39:49 -05:00
kijai
e81f33508b more robust GGUF node detection 2025-09-19 02:05:04 +03:00
kijai
3765985982 Update pyproject.toml 2025-09-18 22:05:38 +03:00
kijai
27c9a4a1cc Update __init__.py 2025-09-18 21:56:42 +03:00
kijai
00da191063 Add BlockifyMask 2025-09-18 21:55:59 +03:00
kijai
edd7994b74 GGUF loader fixes 2025-09-18 21:29:23 +03:00
kijai
623b5913dc bump version 2025-09-18 18:16:11 +03:00
kijai
9f7cf610da Update model_optimization_nodes.py 2025-09-18 18:15:35 +03:00
kijai
8d7a3a3bab Add DrawMaskOnImage 2025-09-18 18:05:34 +03:00
kijai
c2712d7781 Update model_optimization_nodes.py 2025-09-17 17:53:09 +03:00
kijai
39d152e042 Add GGUFLoaderKJ
GGUF loader to extend ComfyUI-GGUF -nodes, requires it installed.
Allows loading separate VACE GGUF modules
2025-09-17 17:50:18 +03:00
kijai
ffd4d1c908 Merge branch 'pr/331' 2025-09-16 18:23:24 +03:00
kijai
1fff7fed6b Rename 2025-09-16 18:23:09 +03:00
Jukka Seppänen
af20cb3333
Merge pull request #375 from asaf92/fix-image-pad-outpaint-mask-scaling
Fix ImagePadForOutpaintTargetSize mask scaling bug
2025-09-16 18:15:28 +03:00
kijai
1885177e8f Merge branch 'pr/359' 2025-09-16 18:14:30 +03:00
kijai
9b43565d38 Move pillarbox_blur mode to ImagePadKJ -node 2025-09-16 18:14:01 +03:00
kijai
5f0a84540b Update image_nodes.py 2025-09-16 17:41:35 +03:00
kijai
468fcc86f0 Add transformer_options for enhance-a-video and NAG patches 2025-09-14 11:49:04 +03:00
kijai
e833a3f7df Fix qwen sage patch too 2025-09-13 12:59:35 +03:00
kijai
5dc5a15cc4 Add transformer_options to sageattn patch 2025-09-13 11:04:17 +03:00
kijai
36f6fdd7d4 Update lora_nodes.py 2025-09-08 10:43:50 +03:00
kijai
6eb25e6b3d Update lora_nodes.py 2025-09-08 10:28:05 +03:00
kijai
142ea4650b Fix logging on lora resize 2025-09-08 10:25:27 +03:00
kijai
0adab07d1a Update lora_nodes.py 2025-09-01 19:02:21 +03:00
kijai
ee5589c9ea Update lora_nodes.py 2025-09-01 18:31:29 +03:00
kijai
d8b8c637fb Add LoraReduceRank
for testing
based on:
https://github.com/kohya-ss/sd-scripts/blob/main/networks/resize_lora.py
2025-09-01 17:11:35 +03:00
aagami
20efb15b44 cleanup 2025-08-25 15:40:12 +03:00
aagami
bc234589e6 Fix ImagePadForOutpaintTargetSize mask scaling bug
When image and mask have different input dimensions and no downscaling
is needed (scaling_factor >= 1), the mask was not being resized to match
the image dimensions before padding, causing mismatched output sizes.

Now ensures mask dimensions match image dimensions using F.interpolate
with nearest-neighbor interpolation in both scaling branches.

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-08-25 14:32:40 +03:00
kijai
ba9153cb06 Allow disabling clamping on lora extract 2025-08-23 16:21:30 +03:00
kijai
e2ce0843d1 bump version 2025-08-20 21:07:03 +03:00
kijai
b37122a9fd Add LazySwitchKJ 2025-08-20 21:06:39 +03:00
kijai
6dfd2c2420 I have no idea... 2025-08-20 14:35:18 +03:00