andreas/ComfyUI
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
3,456 Commits 1 Branch 0 Tags
235d3901fc3f97698e97ff52f61a7caa5f1f11ed
Commit Graph
55 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
comfyanonymousandGitHub d42613686f Fix issue with fp8 ops on some models. (#8045) 
_scaled_mm errors when an input is non contiguous.
 2025-05-10 07:52:56 -04:00
comfyanonymousandGitHub 8ab15c863c Add --mmap-torch-files to enable use of mmap when loading ckpt/pt (#8021) 2025-05-09 04:52:47 -04:00
comfyanonymousandGitHub 924d771e18 Add ACE Step to README. (#8005) 2025-05-08 08:40:57 -04:00
comfyanonymousandGitHub a692c3cca4 Make ACE VAE tiling work. (#8004) 2025-05-08 07:25:45 -04:00
comfyanonymousandGitHub 5d3cc85e13 Make japanese hiragana and katakana characters work with ACE. (#7997) 2025-05-08 03:32:36 -04:00
comfyanonymousandGitHub c7c025b8d1 Adjust memory estimation code for ACE VAE. (#7990) 2025-05-08 01:22:23 -04:00
comfyanonymousandGitHub fd08e39588 Make torchaudio not a hard requirement. (#7987) 
Some platforms can't install it apparently so if it's not there it should
only break models that actually use it.
 2025-05-07 21:37:12 -04:00
comfyanonymousandGitHub 56b6ee6754 Detection code to make ltxv models without config work. (#7986) 2025-05-07 21:28:24 -04:00
comfyanonymousandGitHub cc33cd3422 Experimental lyrics strength for ACE. (#7984) 2025-05-07 19:22:07 -04:00
comfyanonymousandGitHub b9980592c4 Refuse to load api nodes on old pyav version. (#7981) 2025-05-07 17:27:16 -04:00
comfyanonymousandGitHub 16417b40d9 Initial ACE-Step model implementation. (#7972) 2025-05-07 08:33:34 -04:00
comfyanonymousandGitHub 271c9c5b9e Better mem estimation for the LTXV 13B model. (#7963) 2025-05-06 09:52:37 -04:00
comfyanonymousandGitHub a4e679765e Change chroma to use Flux shift. (#7961) 2025-05-06 09:00:01 -04:00
comfyanonymousandGitHub 094e9ef126 Add a way to disable api nodes: --disable-api-nodes (#7960) 2025-05-06 04:53:53 -04:00
comfyanonymousandGitHub d9c80a85e5 This should not be a warning. (#7946) 2025-05-05 07:49:07 -04:00
comfyanonymousandGitHub 80a44b97f5 Change lumina to native RMSNorm. (#7935) 2025-05-04 06:39:23 -04:00
comfyanonymousandGitHub 9187a09483 Change cosmos and hydit models to use the native RMSNorm. (#7934) 2025-05-04 06:26:20 -04:00
comfyanonymousandGitHub 3041e5c354 Switch mochi and wan modes to use pytorch RMSNorm. (#7925) 
* Switch genmo model to native RMSNorm.

* Switch WAN to native RMSNorm.
 2025-05-03 19:07:55 -04:00
comfyanonymousandGitHub 486ad8fdc5 Fix updater issue with newer portable. (#7917) 2025-05-03 00:28:10 -04:00
comfyanonymousandGitHub ff99861650 Make clipsave work with more TE models. (#7908) 2025-05-02 05:15:32 -04:00
comfyanonymousandGitHub aa9d759df3 Switch ltxv to use the pytorch RMSNorm. (#7897) 2025-05-01 06:33:42 -04:00
comfyanonymousandGitHub 39c27a3705 Add updater test to stable release workflow. (#7887) 2025-04-30 14:42:18 -04:00
comfyanonymousandGitHub b1c7291569 Test updater in the windows release workflow. (#7886) 2025-04-30 14:18:20 -04:00
comfyanonymousandGitHub dbc726f80c Better vace memory estimation. (#7875) 2025-04-29 20:42:00 -04:00
comfyanonymousandGitHub 7ee96455e2 Bump minimum pyav version to 14.2.0 (#7874) 2025-04-29 20:38:45 -04:00
comfyanonymousandGitHub 0a66d4b0af Per device stream counters for async offload. (#7873) 2025-04-29 20:28:52 -04:00
comfyanonymousandGitHub 005a91ce2b Latest desktop and portable should work on blackwell. (#7861) 
Removed the mention about the cards from the readme.
 2025-04-29 06:29:38 -04:00
comfyanonymousandGitHub 83d04717b6 Support HiDream E1 model. (#7857) 2025-04-28 15:01:15 -04:00
comfyanonymousandGitHub 5a50c3c7e5 Fix stream priority to support older pytorch. (#7856) 2025-04-28 13:07:21 -04:00
comfyanonymousandGitHub c8cd7ad795 Use stream for casting if enabled. (#7833) 2025-04-27 05:38:11 -04:00
comfyanonymousandGitHub 542b4b36b6 Prevent custom nodes from hooking certain functions. (#7825) 2025-04-26 20:52:56 -04:00
comfyanonymousandGitHub ac10a0d69e Make loras work with --async-offload (#7824) 2025-04-26 19:56:22 -04:00
comfyanonymousandGitHub 0dcc75ca54 Add experimental --async-offload lowvram weight offloading. (#7820) 
This should speed up the lowvram mode a bit. It currently is only enabled when --async-offload is used but it will be enabled by default in the future if there are no problems.
 2025-04-26 16:11:21 -04:00
comfyanonymousandGitHub b685b8a4e0 Update portable package workflow to cu128 (#7812) 2025-04-26 04:43:12 -04:00
comfyanonymousandGitHub 23e39f2ba7 Add a T5TokenizerOptions node to set options for the T5 tokenizer. (#7803) 2025-04-25 19:36:00 -04:00
comfyanonymousandGitHub 5acb705857 Switch LTXVPreprocess to libx264 (#7776) 2025-04-24 13:58:31 -04:00
comfyanonymousandGitHub 154f2911aa Lower size of release package more. (#7754) 2025-04-23 06:33:09 -04:00
comfyanonymousandGitHub 3eaad0590e Lower size of release package. (#7751) 2025-04-23 05:47:09 -04:00
comfyanonymousandGitHub 7eaff81be1 fp16 accumulation can now be enabled on the stable package. (#7750) 2025-04-23 05:28:24 -04:00
comfyanonymousandGitHub 21a11ef817 Pytorch stable 2.7 is out and support cu128 (#7749) 2025-04-23 05:12:59 -04:00
comfyanonymousandGitHub 552615235d Fix for dino lowvram. (#7748) 2025-04-23 04:12:52 -04:00
comfyanonymousandGitHub 2d6805ce57 Add option for using fp8_e8m0fnu for model weights. (#7733) 
Seems to break every model I have tried but worth testing?
 2025-04-22 06:17:38 -04:00
comfyanonymousandGitHub 3ab231f01f Fix issue with WAN VACE implementation. (#7724) 2025-04-21 23:36:12 -04:00
comfyanonymousandGitHub 5d0d4ee98a Add strength control for vace. (#7717) 2025-04-21 19:36:20 -04:00
comfyanonymousandGitHub ce22f687cc Support for WAN VACE preview model. (#7711) 
* Support for WAN VACE preview model.

* Remove print.
 2025-04-21 14:40:29 -04:00
comfyanonymousandGitHub 11b72c9c55 CLIPTextEncodeHiDream. (#7703) 2025-04-21 02:41:51 -04:00
comfyanonymousandGitHub 2c735c13b4 Slightly better fix for #7687 2025-04-20 11:33:27 -04:00
comfyanonymousandGitHub 8c6b9f4481 Prevent custom nodes from accidentally overwriting global modules. (#7167) 
* Prevent custom nodes from accidentally overwriting global modules.

* Improve.
 2025-04-09 09:08:57 -04:00
comfyanonymousandGitHub bc219a6487 Merge pull request #7143 from christian-byrne/fix-remote-widget-node 
Fix LoadImageOutput node
 2025-03-11 04:30:25 -04:00
comfyanonymousandGitHub 94689766ad Merge pull request #7179 from comfyanonymous/ignore_fe_package 
Only check frontend package if using default frontend
 2025-03-11 03:45:02 -04:00
comfyanonymousandGitHub 79d5ceae6e Improved memory management. (#5450) 
* Less fragile memory management.

* Fix issue.

* Remove useless function.

* Prevent and detect some types of memory leaks.

* Run garbage collector when switching workflow if needed.

* Fix issue.
 2024-12-02 14:39:34 -05:00
comfyanonymousandGitHub 6730f3e1a3 Disable non blocking. 
It fixed some perf issues but caused other issues that need to be debugged.
 2024-08-18 14:38:09 -04:00
comfyanonymousandGitHub c032b11e07 xlabs Flux controlnet implementation. (#4260) 
* xlabs Flux controlnet.

* Fix not working on old python.

* Remove comment.
 2024-08-12 21:22:22 -04:00
comfyanonymousandGitHub a5f4292f9f Basic hunyuan dit implementation. (#4102) 
* Let tokenizers return weights to be stored in the saved checkpoint.

* Basic hunyuan dit implementation.

* Fix some resolutions not working.

* Support hydit checkpoint save.

* Init with right dtype.

* Switch to optimized attention in pooler.

* Fix black images on hunyuan dit.
 2024-07-25 18:21:08 -04:00
comfyanonymousGitHubmissionfloyd
 		35a2c790b6 Update comfy_extras/nodes_mask.py 
Co-authored-by: missionfloyd <missionfloyd@users.noreply.github.com>
 2023-04-14 00:12:15 -04:00