comfyanonymous/ComfyUI
1
0
Fork 0
mirror of https://github.com/comfyanonymous/ComfyUI.git synced 2026-02-18 22:20:03 +00:00
Code Issues Packages Projects Releases Wiki Activity
4,739 Commits 106 Branches 121 Tags
aac40e4fca5bd2b87a3d709fce1023bf7f19b39b
Commit Graph

4739 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Jedrzej Kosinski
aac40e4fca Skip replacements that lead to a missing node_id anyway 2026-02-12 22:29:15 -08:00
Jedrzej Kosinski
5b4ba294a1 Merge branch 'master' into jk/node-replace-api 2026-02-12 22:25:03 -08:00
Jedrzej Kosinski
23338b5382 Move node replacement registration for core nodes into nodes_replacements.py 2026-02-12 22:23:26 -08:00
Jedrzej Kosinski
14184a0918 Add apply_replacements to NodeReplaceManager to apply registered node replacements when executing /prompt endpoint 2026-02-12 22:04:52 -08:00
comfyanonymous
e03fe8b591 Update command to install AMD stable linux pytorch. (#12437) 2026-02-12 23:29:12 -05:00
Jedrzej Kosinski
e50ab676ae Merge branch 'master' into jk/node-replace-api 2026-02-12 19:49:54 -08:00
rattus
ae79e33345 llama: use a more efficient rope implementation (#12434) 
Get rid of the cat and unary negation and inplace add-cmul the two
halves of the rope. Precompute -sin once at the start of the model
rather than every transformer block.

This is slightly faster on both GPU and CPU bound setups.
 2026-02-12 19:56:42 -05:00
rattus
117e214354 ModelPatcherDynamic: force load non leaf weights (#12433) 
The current behaviour of the default ModelPatcher is to .to a model
only if its fully loaded, which is how random non-leaf weights get
loaded in non-LowVRAM conditions.

The however means they never get loaded in dynamic_vram. In the
dynamic_vram case, force load them to the GPU.
 2026-02-12 19:51:50 -05:00
Jin Yi
e2f7eaff26 Merge branch 'master' into jk/node-replace-api 2026-02-12 18:40:47 +09:00
Alexander Piskun
4a93a62371 fix(api-nodes): add separate retry budget for 429 rate limit responses (#12421) 2026-02-12 01:38:51 -08:00
comfyanonymous
66c18522fb Add a tip for common error. (#12414) 2026-02-11 22:12:16 -05:00
askmyteapot
e5ae670a40 Update ace15.py to allow min_p sampling (#12373) 2026-02-11 20:28:48 -05:00
rattus
3fe61cedda model_patcher: guard against none model_dtype (#12410) 
Handle the case where the _model_dtype exists but is none with the
intended fallback.
 2026-02-11 14:54:02 -05:00
rattus
2a4328d639 ace15: Use dynamic_vram friendly trange (#12409) 
Factor out the ksampler trange and use it in ACE LLM to prevent the
silent stall at 0 and rate distortion due to first-step model load.
 2026-02-11 14:53:42 -05:00
rattus
d297a749a2 dynamic_vram: Fix windows Aimdo crash + Fix LLM performance (#12408) 
* model_management: lazy-cache aimdo_tensor

These tensors cosntructed from aimdo-allocations are CPU expensive to
make on the pytorch side. Add a cache version that will be valid with
signature match to fast path past whatever torch is doing.

* dynamic_vram: Minimize fast path CPU work

Move as much as possible inside the not resident if block and cache
the formed weight and bias rather than the flat intermediates. In
extreme layer weight rates this adds up.
 2026-02-11 14:50:16 -05:00
Alexander Piskun
2b7cc7e3b6 [API Nodes] enable Magnific Upscalers (#12179) 
* feat(api-nodes): enable Magnific Upscalers

* update price badges

---------

Co-authored-by: Jedrzej Kosinski <kosinkadink1@gmail.com>
 2026-02-11 11:30:19 -08:00
Benjamin Lu
4993411fd9 Dispatch desktop auto-bump when a ComfyUI release is published (#12398) 
* Dispatch desktop auto-bump on ComfyUI release publish

* Fix release webhook secret checks in step conditions

* Require desktop dispatch token in release webhook

* Apply suggestion from @Copilot

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

---------

Co-authored-by: Luke Mino-Altherr <lminoaltherr@gmail.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Jedrzej Kosinski <kosinkadink1@gmail.com>
 2026-02-11 11:15:13 -08:00
Alexander Piskun
2c7cef4a23 fix(api-nodes): retry on connection errors during polling instead of aborting (#12393) 2026-02-11 10:51:49 -08:00
Jedrzej Kosinski
b1c69ed6f6 Improve NodeReplace docstring 2026-02-11 01:16:39 -08:00
Jedrzej Kosinski
1ef4c6e529 Merge branch 'master' into jk/node-replace-api 2026-02-11 01:06:10 -08:00
Jedrzej Kosinski
9e758b5b0c Refactored _node_replace.py InputMap/OutputMap to use a TypedDict instead of objects, simplified the schema sent to the frontend, updated nodes_post_processing.py replacements to use new schema 2026-02-11 01:02:55 -08:00
comfyanonymous
76a7fa96db Make built in lora training work on anima. (#12402) 2026-02-10 22:04:32 -05:00
Kohaku-Blueleaf
cdcf4119b3 [Trainer] training with proper offloading (#12189) 
* Fix bypass dtype/device moving

* Force offloading mode for training

* training context var

* offloading implementation in training node

* fix wrong input type

* Support bypass load lora model, correct adapter/offloading handling
 2026-02-10 21:45:19 -05:00
Jedrzej Kosinski
a6d691dc45 Merge branch 'master' into jk/node-replace-api 2026-02-10 16:54:35 -06:00
Christian Byrne
8d0da49499 feat: add node_replacements server feature flag (#12362) 
Amp-Thread-ID: https://ampcode.com/threads/T-019c3f3d-e208-704f-bf25-4f643c1e0059
 2026-02-10 14:53:28 -08:00
AustinMroz
dbe70b6821 Add a VideoSlice node (#12107) 
* Base TrimVideo implementation

* Raise error if as_trimmed call fails

* Bigger max start_time, tooltips, and formatting

* Count packets unless codec has subframes

* Remove incorrect nested decode

* Add null check for audio streams

* Support non-strict duration

* Added strict_duration bool to node definition

* Empty commit for approval

* Fix duration

* Support 5.1 audio layout on save

---------

Co-authored-by: Jedrzej Kosinski <kosinkadink1@gmail.com>
 2026-02-10 14:42:21 -08:00
guill
00fff6019e feat(jobs): add 3d to PREVIEWABLE_MEDIA_TYPES for first-class 3D output support (#12381) 
Co-authored-by: Jedrzej Kosinski <kosinkadink1@gmail.com>
 2026-02-10 14:37:14 -08:00
rattus
123a7874a9 ops: Fix vanilla-fp8 loaded lora quality (#12390) 
This was missing the stochastic rounding required for fp8 downcast
to be consistent with model_patcher.patch_weight_to_device.

Missed in testing as I spend too much time with quantized tensors
and overlooked the simpler ones.
 2026-02-10 13:38:28 -05:00
rattus
f719f9c062 sd: delay VAE dtype archive until after override (#12388) 
VAEs have host specific dtype logic that should override the dynamic
_model_dtype. Defer the archiving of model dtypes until after.
 2026-02-10 13:37:46 -05:00
rattus
fe053ba5eb mp: dont deep-clone objects from model_options (#12382) 
If there are non-trivial python objects nested in the model_options, this
causes all sorts of issues. Traverse lists and dicts so clones can safely
overide settings and BYO objects but stop there on the deepclone.
 2026-02-10 13:37:17 -05:00
comfyanonymous
6648ab68bc ComfyUI v0.13.0 v0.13.0 2026-02-10 13:26:29 -05:00
ComfyUI Wiki
6615db925c chore: update workflow templates to v0.8.38 (#12394) 2026-02-10 13:24:56 -05:00
Alexander Piskun
8ca842a8ed feat(api-nodes-Kling): add new models (V3, O3) (#12389) 
* feat(api-nodes-Kling): add new models (V3, O3)

* remove storyboard from VideoToVideo node

* added check for total duration of storyboards

* fixed other small things

* updated display name for nodes

* added "fake" seed
 2026-02-10 09:34:54 -08:00
Alexander Piskun
c1b63a7e78 fix(Moonvalley-API-Nodes): adjust "steps" parameter to not raise exception (#12370) 2026-02-09 21:58:27 -05:00
ComfyUI Wiki
349a636a2b chore: update workflow templates to v0.8.37 (#12377) 2026-02-09 21:25:34 -05:00
comfyanonymous
a4be04c5d7 Ace step prompts match now. (#12376) 2026-02-09 19:45:56 -05:00
blepping
baf8c87455 Iimprovements to ACE-Steps 1.5 text encoding (part 2) (#12350) 2026-02-09 19:41:49 -05:00
rattus
62315fbb15 Dynamic VRAM fixes - Ace 1.5 performance + a VRAM leak (#12368) 
* revert threaded model loader change

This change was only needed to get around the pytorch 2.7 mempool bugs,
and should have been reverted along with #12260. This fixes a different
memory leak where pytorch gets confused about cache emptying.

* load non comfy weights

* MPDynamic: Pre-generate the tensors for vbars

Apparently this is an expensive operation that slows down things.

* bump to aimdo 1.8

New features:
watermark limit feature
logging enhancements
-O2 build on linux
 2026-02-09 16:16:08 -05:00
comfyanonymous
a0302cc6a8 Make tonemap latent work on any dim latents. (#12363) 2026-02-08 21:16:40 -05:00
comfyanonymous
f350a84261 Disable prompt weights for ltxv2. (#12354) 2026-02-07 19:16:28 -05:00
ComfyUI Wiki
3760d74005 chore: update embedded docs to v0.4.1 (#12346) 2026-02-07 18:34:52 -05:00
chaObserv
9bf5aa54db Add search_aliases to sa-solver and seeds-2 node (#12327) 2026-02-07 17:38:51 -05:00
Jukka Seppänen
5ff4fdedba Fix LazyCache (#12344) 2026-02-07 11:25:30 -08:00
comfyanonymous
17e7df43d1 Pad ace step 1.5 ref audio if not long enough. (#12341) 2026-02-07 00:02:11 -05:00
comfyanonymous
039955c527 Some fixes to previous pr. (#12339) 2026-02-06 20:14:52 -05:00
tdrussell
6a26328842 Support fp16 for Cosmos-Predict2 and Anima (#12249) 2026-02-06 20:12:15 -05:00
comfyanonymous
204e65b8dc Fix bug with last pr (#12338) 2026-02-06 19:48:20 -05:00
asagi4
a831c19b70 Fix return_word_ids=True with Anima tokenizer (#12328) 2026-02-06 19:38:04 -05:00
comfyanonymous
eba6c940fd Make ace step 1.5 base model work properly with default workflow. (#12337) 2026-02-06 19:14:56 -05:00
Jukka Seppänen
a1c101f861 EasyCache: Support LTX2 (#12231) 2026-02-06 00:43:09 -05:00