Commit Graph

Select branches

Hide Pull Requests

bindweave

causvid

longcat_avatar

longvie2

main

self-refine-video

syncammaster

vap

Mono Color

• 3183589eff Merge branch 'main' of https://github.com/kijai/ComfyUI-WanVideoWrapper kijai 2025-09-02 17:09:12 +03:00
• 4cf23a5a40 Can't do this inplace with all models kijai 2025-09-02 17:09:05 +03:00
• 0405dde5a2 Merge pull request #1165 from zinigor/fix/unbound-local-error Jukka Seppänen 2025-09-02 15:06:44 +03:00
• 42054724b7 Moved dimension update to where the count is initialized. Igor Zinovyev 2025-09-02 14:32:01 +03:00
• 42d7180759 keep scale_weight on gpu to avoid needless casts kijai 2025-09-02 01:04:16 +03:00
• 82608009c8 use a copy of the scheduler if provided by the scheduler node to make sure it's always reset kijai 2025-09-01 23:04:05 +03:00
• f9755820de handle negative latent indices with extra latents kijai 2025-09-01 14:19:51 +03:00
• 56e3366c26 Fix pusa steps kijai 2025-09-01 13:50:41 +03:00
• 43d8a2c8a5 flowmatch_pusa split sampling fixes kijai 2025-09-01 13:34:49 +03:00
• d4d942c5ee Update nodes.py kijai 2025-09-01 11:17:22 +03:00
• 30c2fca799 WanVideoScheduler fixes kijai 2025-09-01 11:07:41 +03:00
• d9038d575f Allow setting both start and end step by sigma kijai 2025-08-31 19:04:01 +03:00
• b8acd3ee48 Allow using fp16 multitalk model with GGUF again kijai 2025-08-31 18:57:59 +03:00
• 9f84cd2254 Fix offloading when using stand-in lora kijai 2025-08-31 18:45:13 +03:00
• b9b46638ac Fix FantasyPortrait + GGUF main model kijai 2025-08-31 18:41:08 +03:00
• b1c8b8a280 remove flex attention as redundant kijai 2025-08-31 18:14:38 +03:00
• 79395b872c Update pyproject.toml kijai 2025-08-31 18:10:07 +03:00
• cedac626e2 Update __init__.py kijai 2025-08-31 18:09:24 +03:00
• 4870b8f081 Update __init__.py kijai 2025-08-31 18:06:20 +03:00
• 81dac31735 Merge branch 's2v' kijai 2025-08-31 18:05:30 +03:00
• 26a11c3044 Update nodes.py kijai 2025-08-31 18:01:53 +03:00
• 10d3a0fbef Update model.py kijai 2025-08-30 23:34:14 +03:00
• 40308f1c70 Allow basic FantasyPortrait + S2V kijai 2025-08-30 22:33:32 +03:00
• df0b8c419a Fix kijai 2025-08-30 18:01:05 +03:00
• 45f129b287 Revert this for now to avoid issues on some systems kijai 2025-08-30 14:09:16 +03:00
• a7cf6d4ed1 Update nodes.py kijai 2025-08-29 21:33:12 +03:00
• 7dd0e1ac61 Update nodes_model_loading.py kijai 2025-08-29 21:21:24 +03:00
• f36fa45c29 Loading fixes when using merged loras kijai 2025-08-29 20:57:28 +03:00
• 3a290cfd15 Update nodes_model_loading.py kijai 2025-08-29 20:41:40 +03:00
• 8f6507fa64 Update nodes.py kijai 2025-08-29 18:10:18 +03:00
• 1e4cc84a30 Fix previous kijai 2025-08-29 17:35:27 +03:00
• 88cf36936b skip this if not on cuda kijai 2025-08-29 17:29:25 +03:00
• 876b5bda5c Update nodes.py kijai 2025-08-29 17:17:37 +03:00
• 5f7a5d533b Reduce memory use in the Framepack loop kijai 2025-08-29 16:59:23 +03:00
• 6a2053c9d1 Update nodes_model_loading.py kijai 2025-08-29 15:26:47 +03:00
• 0c8a883a36 Update nodes.py kijai 2025-08-29 15:25:31 +03:00
• 044ff9f25e don't use model management for cuda stream... kijai 2025-08-29 15:21:49 +03:00
• 6978686272 Update model.py kijai 2025-08-29 15:16:02 +03:00
• 740157bd61 Fix MultiTalk when using the Multi/InfiniteTalk node kijai 2025-08-29 14:48:12 +03:00
• 704cca215b Fix control lora when not last lora to be loaded kijai 2025-08-29 14:45:03 +03:00
• 999a314652 Fix GGUF offloading kijai 2025-08-29 12:58:36 +03:00
• f9be754980 cleanup kijai 2025-08-29 00:17:05 +03:00
• a21e4b3210 Update model.py kijai 2025-08-28 20:54:36 +03:00
• 469d9168da Update wanvideo2_2_S2V_framepack_pose_testing.json kijai 2025-08-28 19:31:47 +03:00
• a053336c1a Update nodes.py kijai 2025-08-28 18:52:10 +03:00
• 6faa24b7f2 Implement Framepack long geneneration method kijai 2025-08-28 18:48:39 +03:00
• a5621b8739 Add pose input kijai 2025-08-27 22:43:28 +03:00
• 5266959a93 Create wanvideo2_2_S2V_context_window_testing.json kijai 2025-08-27 21:41:34 +03:00
• a37b12235f Add loudness norm node kijai 2025-08-27 20:51:08 +03:00
• 90c3bbb6c2 better context window indices, cleanup kijai 2025-08-27 18:47:31 +03:00
• 4bc1ffee24 Update wanvideo_Fun_2_2_control_example_03.json kijai 2025-08-27 16:12:31 +03:00
• 56bd2ec5f6 Update WanVideoScheduler -node kijai 2025-08-27 13:19:07 +03:00
• a507a5763d update WanVideoScheduler -node kijai 2025-08-27 13:03:24 +03:00
• c89b82e479 Merge branch 'main' into s2v kijai 2025-08-27 12:24:08 +03:00
• e241998f53 Fix Uni3C kijai 2025-08-27 12:21:25 +03:00
• 0538f5d600 Update nodes.py kijai 2025-08-27 02:41:24 +03:00
• a84d142cb2 Update nodes.py kijai 2025-08-27 02:40:27 +03:00
• 5d17484cc5 continue kijai 2025-08-27 02:25:39 +03:00
• 636b252c7e Merge branch 'main' into s2v kijai 2025-08-26 22:41:37 +03:00
• 2a3ca69c83 Fix offloading when using merged loras kijai 2025-08-26 22:34:12 +03:00
• e85e1ff9e6 Update model.py kijai 2025-08-26 22:05:22 +03:00
• 8e65fae3c5 Update model.py kijai 2025-08-26 22:02:36 +03:00
• 96f7f6accd fix normal model loading kijai 2025-08-26 22:02:02 +03:00
• 3c79851230 ref latent kijai 2025-08-26 21:30:10 +03:00
• 63d4b6aada Update nodes.py kijai 2025-08-26 19:07:38 +03:00
• 748ec89aa8 init kijai 2025-08-26 19:02:24 +03:00
• a1ca0985ec Merge branch 'dev' kijai 2025-08-26 16:39:26 +03:00
• 30f185a19b Update pyproject.toml kijai 2025-08-26 16:39:17 +03:00
• 4f13aa46d4 Update nodes.py kijai 2025-08-26 01:12:22 +03:00
• dfb2a59211 Update nodes.py kijai 2025-08-26 01:11:20 +03:00
• 5f5f8ae30a Merge branch 'main' into dev kijai 2025-08-26 00:33:49 +03:00
• 82d2ab44e5 update workflows kijai 2025-08-26 00:27:08 +03:00
• efcef40b00 fix name kijai 2025-08-26 00:18:05 +03:00
• 6fce0e2d3b Add Wav2VecModelLoader to load wav2vec2 from single .safetensors kijai 2025-08-26 00:09:46 +03:00
• 665ddd6811 Merge branch 'main' into dev kijai 2025-08-25 19:35:13 +03:00
• e836134b90 UniAnimate fixes and optimization kijai 2025-08-25 19:35:05 +03:00
• 17f02a91bf Merge branch 'main' into dev kijai 2025-08-25 17:29:04 +03:00
• 4d52c10148 Faster Multi/InfiniteTalk model load kijai 2025-08-25 15:06:31 +03:00
• 9c63d8d04f Fix temporal_mask dtype kijai 2025-08-25 14:59:43 +03:00
• 288a7e96a9 Add experimental scheduler node with sigma plot kijai 2025-08-25 14:53:36 +03:00
• d9def84332 Fix UniAnimate and MultiTalk loading kijai 2025-08-25 14:09:08 +03:00
• 42df277488 Merge branch 'main' into dev kijai 2025-08-25 12:51:46 +03:00
• 66d4c5b90d Slice possible alpha channel away before encoding kijai 2025-08-25 12:50:41 +03:00
• bc22008b66 Fix InfiniteTalk v2v kijai 2025-08-24 22:58:10 +03:00
• c65f81d089 Handle missing faces better in FantasyPortrait detection kijai 2025-08-24 20:50:11 +03:00
• 71c93f87fd Update nodes.py kijai 2025-08-24 20:28:26 +03:00
• 0c3030a7d8 Add a node to easier split sigmas, EasyCache fixes kijai 2025-08-24 20:07:34 +03:00
• 3d9b52cb10 Fix: can't del the end image at this point... kijai 2025-08-24 14:18:54 +03:00
• 61aa0e2ffd Drastically reduce VRAM usage on I2V encode node. kijai 2025-08-24 11:39:56 +03:00
• 8a151b5402 Multi/InfiniteTalk sampling loop cleanup and optimizations, support FantasyPortrait within the loop kijai 2025-08-24 11:39:18 +03:00
• 54f745bb63 Update nodes.py Jukka Seppänen 2025-08-24 02:03:32 +03:00
• d39155213e This can be float for compatibility kijai 2025-08-24 01:06:53 +03:00
• 9b98413e1e Update nodes.py kijai 2025-08-23 21:00:48 +03:00
• 11f3e41e9a Merge branch 'main' into dev kijai 2025-08-23 21:00:45 +03:00
• 9076a3aae8 bump version kijai 2025-08-23 20:57:41 +03:00
• b7d7f9afe5 cleanup unianimate stuff kijai 2025-08-23 18:59:21 +03:00
• 77d79f6878 This is redundant kijai 2025-08-23 17:55:12 +03:00
• 9a2a13498a Cleanup MultiTalk model code kijai 2025-08-23 17:49:04 +03:00
• 73cff6ebab Cleanup Multi/InfiniteTalk sampling loop some kijai 2025-08-23 16:27:10 +03:00
• 5f521dc169 Add generic CreateScheduleFloatList to assist with LoRA scheduling kijai 2025-08-23 13:04:19 +03:00