1
0
mirror of https://github.com/kijai/ComfyUI-WanVideoWrapper.git synced 2026-01-26 23:41:35 +03:00

Commit Graph

  • 369043c5d8 Add comfy core attention as option kijai 2025-12-10 19:39:05 +02:00
  • eb6dd96049 Merge pull request #1749 from wong00/fix-flashattn3-attention-compute Jukka Seppänen 2025-12-10 11:52:18 +02:00
  • 38a48c670a [bugfix] Fixed the logic for processing return values after computing attention in flashAttention3 wangxin68 2025-12-10 17:27:37 +08:00
  • dd58511d4e Expose some WanMove track preview drawing options kijai 2025-12-10 02:26:51 +02:00
  • 4c97d27583 Fix LongCat error kijai 2025-12-09 21:09:44 +02:00
  • bcdc0c0661 Add node to use WanMove with native workflows kijai 2025-12-09 21:04:56 +02:00
  • 83c25644ef Create wanvideo_WanMove_I2V_example_01.json kijai 2025-12-09 20:28:45 +02:00
  • 113b6df04d Fix multiple tracks kijai 2025-12-09 20:11:45 +02:00
  • d000fbc645 Support WanMove kijai 2025-12-09 19:55:56 +02:00
  • cc58964027 Refactor init.py to something more sensible kijai 2025-12-09 19:53:26 +02:00
  • fa57681424 Reduce recompiles with unmerged loras kijai 2025-12-09 14:40:52 +02:00
  • 0a65354247 Allow larger wananimate window size for disabling windowing kijai 2025-12-09 14:15:00 +02:00
  • a27a4892b6 version 1.4.1 kijai 2025-12-09 12:56:40 +02:00
  • 8b037bce2e Squashed commit of the following: kijai 2025-12-09 12:56:11 +02:00
  • e867e642d4 Fix longcat scheduler kijai 2025-12-08 10:26:38 +02:00
  • 00971fc121 Don't remove lora in loop sampling offload kijai 2025-12-08 09:43:36 +02:00
  • 0e81adb843 Fix for HuMo with I2V Jukka Seppänen 2025-12-08 03:17:29 +02:00
  • 7f9560eb33 Fix mistake in sageattn_ultravico kijai 2025-12-07 19:56:03 +02:00
  • 2369cdbbe9 Fix prompt splitting kijai 2025-12-07 18:34:21 +02:00
  • 123c9ca312 Allow HuMo to work with start_image kijai 2025-12-07 18:18:16 +02:00
  • 2b62866945 Fix SteadyDancer GGUF dtypes kijai 2025-12-06 17:25:57 +02:00
  • b5415573c5 Update nodes_model_loading.py kijai 2025-12-05 20:55:33 +02:00
  • 9e005fab90 Fix scaling for light TAE kijai 2025-12-05 14:03:33 +02:00
  • f85ea5a48a add "empty_frame_pad_image" input to I2V encode node for easier SVI lora use kijai 2025-12-05 13:05:35 +02:00
  • ee1bb5c5c7 Update nodes_utility.py kijai 2025-12-05 12:07:48 +02:00
  • 041c31fe2e cleanup kijai 2025-12-05 12:01:34 +02:00
  • 4e7b8dd92c Clearer error on attention mode imports kijai 2025-12-05 02:00:31 +02:00
  • e2333d0f04 Update nodes_sampler.py kijai 2025-12-04 16:51:30 +02:00
  • 2dd4c6bf15 Merge branch 'pr/1705' kijai 2025-12-04 16:48:50 +02:00
  • 9d0a21228b Update nodes_sampler.py kijai 2025-12-04 16:48:23 +02:00
  • 001bc0e24a Merge branch 'main' of https://github.com/kijai/ComfyUI-WanVideoWrapper kijai 2025-12-04 16:44:29 +02:00
  • 7189922f36 Update __init__.py kijai 2025-12-04 16:44:18 +02:00
  • faf9c5927d Merge branch 'pr/1510' kijai 2025-12-04 16:44:02 +02:00
  • 6c5dc0ba2a Merge pull request #1546 from wzxysf/main Jukka Seppänen 2025-12-04 16:40:29 +02:00
  • 995804166d Merge pull request #1651 from jamesjjcondon/patch-1 Jukka Seppänen 2025-12-04 16:40:05 +02:00
  • 4e2bf0f1fa Fix device mismatch error with I2V + Lynx embeds mossmatrix 2025-12-04 00:34:54 -05:00
  • b06c7d2d6d Add UltraViCo -sage attention mode and refactor some attention code kijai 2025-12-03 13:32:45 +02:00
  • 014e711972 cleanup kijai 2025-12-03 11:09:09 +02:00
  • c47b1ded69 Apply conv3d workaround to UniAnimate kijai 2025-12-02 15:00:06 +02:00
  • 7bc45daaf2 Cleanup whitespaces kijai 2025-12-02 14:50:26 +02:00
  • aebeeb9160 Code cleanup kijai 2025-12-02 14:49:42 +02:00
  • e60eb995ee version 1.4.0 kijai 2025-12-02 00:46:01 +02:00
  • b9f6c9aa50 Update __init__.py kijai 2025-12-02 00:45:22 +02:00
  • c1fbc93521 Add ViBTScheduler to use ViBT models kijai 2025-12-02 00:26:50 +02:00
  • c4ca252fea Fix compiled LoRA application kijai 2025-12-01 20:38:01 +02:00
  • c4db00609a Use torch.chunk for chunking kijai 2025-12-01 17:39:50 +02:00
  • 5a52d6b92f Add VAE feat_cache offloading, VAE tqdm progress bar and memory usage report kijai 2025-12-01 15:05:20 +02:00
  • a652c55bf7 Remove LoRAs when fully offloading as well kijai 2025-12-01 12:50:22 +02:00
  • 196d39695f Fix sageattn_varlen kijai 2025-12-01 01:11:06 +02:00
  • e5be3e5263 Use torch custom_ops to avoid graph breaks with torch.compile kijai 2025-12-01 00:29:27 +02:00
  • a6071c7be5 Merge branch 'main' into steadydancer kijai 2025-11-30 17:56:50 +02:00
  • 0cba1edd4e Better just not compile this as it's causing issues kijai 2025-11-30 17:53:28 +02:00
  • a9cd073f29 Remove unnecessary recompile when using cfg kijai 2025-11-30 17:52:56 +02:00
  • 1e9e2be622 Avoid recompile here kijai 2025-11-30 17:32:24 +02:00
  • 99c3978da4 Reduce peak VRAM usage when not using torch.compile (and some even with it) kijai 2025-11-30 17:14:53 +02:00
  • 30bd7d46eb example update kijai 2025-11-29 02:38:02 +02:00
  • 0c9d5b8dcc context windows kijai 2025-11-29 02:11:57 +02:00
  • 66d44ec8db This doesn't really do anything useful kijai 2025-11-28 21:15:25 +02:00
  • 394c7c13d2 Add strength controls kijai 2025-11-28 21:11:20 +02:00
  • c9931364b3 Create wanvideo_I2V_steadydancer_testing.json kijai 2025-11-28 20:34:55 +02:00
  • e54fa5d059 Init kijai 2025-11-28 20:32:16 +02:00
  • 772642b4f1 Update nodes.py kijai 2025-11-27 20:19:06 +02:00
  • 472ed70757 Add node to preview image embeds kijai 2025-11-27 20:08:23 +02:00
  • 44feb24290 Allow TTM to work with 5B models kijai 2025-11-20 11:43:53 +02:00
  • fa7a967ee7 Fix stand-in kijai 2025-11-17 11:56:02 +02:00
  • ec161373f4 Update readme.md kijai 2025-11-16 19:30:09 +02:00
  • 0e3fd0b491 Create wanvideo2_2_I2V_A14B_TimeToMove_example.json kijai 2025-11-16 19:26:51 +02:00
  • f872460285 Fix TTM for dual sampler setups kijai 2025-11-16 19:23:19 +02:00
  • b826642a83 Add TTM support (Time To Move) kijai 2025-11-16 17:52:49 +02:00
  • aa9f474958 Refactor data_mean and data_std initialization jamesjjcondon 2025-11-14 11:57:26 +10:30
  • e3c2a1431b Squashed commit of the following: kijai 2025-11-13 17:10:57 +02:00
  • f685ee33ac Merge branch 'main' into bindweave bindweave kijai 2025-11-13 16:37:38 +02:00
  • 4e31081262 Better errors when trying to load models that don't exist kijai 2025-11-13 16:19:05 +02:00
  • bb5707f601 Merge branch 'main' into bindweave kijai 2025-11-11 18:53:19 +02:00
  • ff26836cab Create wanvideo_2_2_5B_Ovi_image_to_video_audio_10_seconds_example_01.json kijai 2025-11-11 18:53:11 +02:00
  • 22037243ab Fix Ovi audio negative prompt kijai 2025-11-11 17:47:33 +02:00
  • acb662b5af Merge branch 'main' into bindweave kijai 2025-11-11 11:44:26 +02:00
  • e926f7a069 version bump 1.3.9 kijai 2025-11-11 10:57:46 +02:00
  • e01e34da1f Update nodes_model_loading.py kijai 2025-11-11 10:01:06 +02:00
  • 47514f678d Allow loading original Ovi -models kijai 2025-11-11 09:46:13 +02:00
  • 907c9e1cdd Update nodes.py kijai 2025-11-10 21:02:58 +02:00
  • de3c9c895a Create wanvideo_1_3B_UniLumos_relight_example_01.json kijai 2025-11-10 19:41:22 +02:00
  • 4576ddb35e Add node to create input for UniLumos kijai 2025-11-10 19:41:19 +02:00
  • 68392684b5 Add node to use UniLumos kijai 2025-11-10 19:06:45 +02:00
  • e4a4d22537 Update nodes_sampler.py kijai 2025-11-08 16:04:55 +02:00
  • a3b2f67337 Pad clip vision embeds like in original code kijai 2025-11-08 16:03:00 +02:00
  • 1e00c8fb28 Update nodes.py kijai 2025-11-08 12:21:11 +02:00
  • ff16dce5c0 Update nodes_sampler.py kijai 2025-11-07 01:15:13 +02:00
  • f972b31bf2 Update nodes.py kijai 2025-11-07 01:09:22 +02:00
  • 3dacd6a719 Update nodes_sampler.py kijai 2025-11-07 00:45:06 +02:00
  • 7bf99791ad Update nodes.py kijai 2025-11-07 00:44:13 +02:00
  • 7a5587b5af Let the user resize for QwenVL kijai 2025-11-07 00:39:10 +02:00
  • d6cf172846 Update nodes.py kijai 2025-11-06 23:41:24 +02:00
  • cf86f4f0a4 Update model.py kijai 2025-11-06 21:01:03 +02:00
  • b1f8309a20 Update nodes_model_loading.py kijai 2025-11-06 19:55:54 +02:00
  • 8992c6af64 Don't include padding for scheduler kijai 2025-11-06 19:22:56 +02:00
  • e4084a961b Update nodes.py kijai 2025-11-06 18:32:53 +02:00
  • 3ec1edefbe init kijai 2025-11-06 17:35:48 +02:00
  • d3f33a9f09 Update readme.md Jukka Seppänen 2025-11-06 16:38:46 +02:00
  • d0ef3b5601 Update readme.md kijai 2025-11-06 16:37:50 +02:00