114fc73685
Bump comfyui-frontend-package to 1.36.13 ( #11645 )
2026-01-08 22:16:15 -05:00
b48d6a83d4
Fix csp error in frontend when forcing offline. ( #11749 )
2026-01-08 22:15:50 -05:00
027042db68
Add node: JoinAudioChannels ( #11728 )
2026-01-08 22:14:06 -05:00
1a20656448
Fix import issue. ( #11746 )
2026-01-08 17:23:59 -05:00
0f11869d55
Better detection if AMD torch compiled with efficient attention. ( #11745 )
2026-01-08 17:16:58 -05:00
5943fbf457
bump comfyui_manager version to the 4.0.5 ( #11732 )
2026-01-08 08:15:42 -08:00
a60b7b86c5
Revert "Force sequential execution in CI test jobs ( #11687 )" ( #11725 )
...
This reverts commit ce0000c4f2
2026-01-07 21:41:57 -08:00
2e9d51680a
ComfyUI version v0.8.2
2026-01-07 23:50:02 -05:00
50d6e1caf4
Tweak ltxv vae mem estimation. ( #11722 )
2026-01-07 23:07:05 -05:00
ac12f77bed
ComfyUI version v0.8.1
2026-01-07 22:10:08 -05:00
fcd9a236b0
Update template to 0.7.69 ( #11719 )
2026-01-07 18:22:23 -08:00
21e8425087
Add warning for old pytorch. ( #11718 )
2026-01-07 21:07:26 -05:00
b6c79a648a
ops: Fix offloading with FP8MM performance ( #11697 )
...
This logic was checking comfy_cast_weights, and going straight to
to the forward_comfy_cast_weights implementation without
attempting to downscale input to fp8 in the event comfy_cast_weights
is set.
The main reason comfy_cast_weights would be set would be for async
offload, which is not a good reason to nix FP8MM.
So instead, and together the underlying exclusions for FP8MM which
are:
* having a weight_function (usually LowVramPatch)
* force_cast_weights (compute dtype override)
* the weight is not Quantized
* the input is already quantized
* the model or layer has MM explictily disabled.
If you get past all of those exclusions, quantize the input tensor.
Then hand the new input, quantized or not off to
forward_comfy_cast_weights to handle it. If the weight is offloaded
but input is quantized you will get an offloaded MM8.
2026-01-07 21:01:16 -05:00
25bc1b5b57
Add memory estimation function to ltxav text encoder. ( #11716 )
2026-01-07 20:11:22 -05:00
3cd19e99c1
Increase ltxav mem estimation by a bit. ( #11715 )
2026-01-07 20:04:56 -05:00
007b87e7ac
Bump required comfy-kitchen version. ( #11714 )
2026-01-07 19:48:47 -05:00
34751fe9f9
Lower ltxv text encoder vram use. ( #11713 )
2026-01-07 19:12:15 -05:00
1c705f7bfb
Add device selection for LTXAVTextEncoderLoader ( #11700 )
2026-01-07 18:39:59 -05:00
48e5ea1dfd
model_patcher: Remove confusing load stat ( #11710 )
...
If the loader passes 1e32 as the usable memory size, it means force
the full load. This happens with CPU loads and a few other misc cases.
Removing the confusing number and just leave the other details.
2026-01-07 18:39:20 -05:00
3cd7b32f1b
Support gemma 12B with quant weights. ( #11696 )
2026-01-07 05:15:14 -05:00
c0c9720d77
Fix stable release workflow not pulling latest comfy kitchen. ( #11695 )
2026-01-07 04:48:28 -05:00
fc0cb10bcb
ComfyUI v0.8.0
2026-01-07 04:07:31 -05:00
b7d7cc1d49
Fix fp8 fast issue. ( #11688 )
2026-01-07 01:39:06 -05:00
79e94544bd
feat(api-nodes): add WAN2.6 ReferenceToVideo ( #11644 )
2026-01-06 22:04:50 -08:00
ce0000c4f2
Force sequential execution in CI test jobs ( #11687 )
...
Added max-parallel setting to enforce sequential execution in test jobs.
2026-01-07 00:57:31 -05:00
c5cfb34c07
Update comfy-kitchen version to 0.2.3 ( #11685 )
2026-01-06 23:51:45 -05:00
edee33f55e
Disable comfy kitchen cuda if pytorch cuda less than 13 ( #11681 )
2026-01-06 22:13:43 -05:00
2c03884f5f
Skip fp4 matrix mult on devices that don't support it. ( #11677 )
2026-01-06 18:07:26 -05:00
6e9ee55cdd
Disable ltxav previews. ( #11676 )
2026-01-06 17:41:27 -05:00
023cf13721
Fix lowvram issue with ltxv2 text encoder. ( #11675 )
2026-01-06 17:33:03 -05:00
c3566c0d76
chore: update workflow templates to v0.7.67 ( #11667 )
2026-01-06 14:28:29 -08:00
c3c3e93c5b
Use rope functions from comfy kitchen. ( #11674 )
2026-01-06 16:57:50 -05:00
6ffc159bdd
Update comfy-kitchen version to 0.2.1 ( #11672 )
2026-01-06 15:53:43 -05:00
96e0d0924e
Add helpful message to portable. ( #11671 )
2026-01-06 14:43:24 -05:00
e14f3b6610
chore: update workflow templates to v0.7.66 ( #11652 )
2026-01-05 22:37:11 -08:00
1618002411
Revert "Use rope functions from comfy kitchen. ( #11647 )" ( #11648 )
...
This reverts commit 6ef85c4915
2026-01-05 23:07:39 -05:00
6ef85c4915
Use rope functions from comfy kitchen. ( #11647 )
2026-01-05 22:50:35 -05:00
6da00dd899
Initial ops changes to use comfy_kitchen: Initial nvfp4 checkpoint support. ( #11635 )
...
---------
Co-authored-by: Jedrzej Kosinski <kosinkadink1@gmail.com >
2026-01-05 21:48:58 -05:00
4f3f9e72a9
Fix name. ( #11638 )
2026-01-05 02:41:23 -08:00
d157c3299d
Refactor module_size function. ( #11637 )
2026-01-05 03:48:31 -05:00
d1b9822f74
Add LTXAVTextEncoderLoader node. ( #11634 )
2026-01-05 02:27:31 -05:00
f2b002372b
Support the LTXV 2 model. ( #11632 )
2026-01-05 01:58:59 -05:00
38d0493825
Fix case where upscale model wouldn't be moved to cpu. ( #11633 )
2026-01-04 19:13:50 -05:00
acbf08cd60
feat(api-nodes): add support for 720p resolution for Kling Omni nodes ( #11604 )
2026-01-03 23:05:02 -08:00
53e762a3af
Print memory summary on OOM to help with debugging. ( #11613 )
2026-01-03 22:28:38 -05:00
9a552df898
Remove leftover scaled_fp8 key. ( #11603 )
2026-01-02 17:28:10 -08:00
f2fda021ab
Tripo3D: pass face_limit parameter only when it differs from default ( #11601 )
2026-01-02 03:18:43 -08:00
303b1735f8
Give Mahiro CFG a more appropriate display name ( #11580 )
2026-01-02 00:37:37 -08:00
9e5f677746
Ignore all frames except the first one for MPO format. ( #11569 )
2026-01-02 00:35:34 -08:00
65cfcf5b1b
New Year ruff cleanup. ( #11595 )
2026-01-01 22:06:14 -05:00
