Commit Graph

  • 7946049794 nit Chenlei Hu 2025-03-10 15:14:40 -04:00
  • 6f6349b6a7 nit Chenlei Hu 2025-03-10 15:10:40 -04:00
  • 1f138dd382 Only check frontend package if using default frontend Chenlei Hu 2025-03-10 15:07:44 -04:00
  • b779349b55 Temporarily revert fix to give time for people to update their nodes. comfyanonymous 2025-03-10 06:30:17 -04:00
  • 35e2dcf5d7 Hack to fix broken manager. comfyanonymous 2025-03-10 06:14:43 -04:00
  • 67c7184b74
    ltxv: relax frame_idx divisibility for single frames. (#7146) Andrew Kvochko 2025-03-10 10:11:48 +02:00
  • 6f8e766509 Prevent custom nodes from accidentally overwriting global modules. comfyanonymous 2025-03-10 03:33:17 -04:00
  • e1da98a14a
    remove unused params (#6931) Terry Jia 2025-03-09 14:07:09 -04:00
  • a73410aafa remove overrides bymyself 2025-03-09 03:46:08 -07:00
  • 9aac21f894 Fix issues with new hunyuan img2vid model and bumb version to v0.3.26 comfyanonymous 2025-03-09 04:59:15 -04:00
  • 528d1b3563
    When cached_hook_patches contain weights for hooks, only use hook_backup for unused keys (#7067) Jedrzej Kosinski 2025-03-09 03:26:31 -05:00
  • 2bc4b5968f ComfyUI version v0.3.25 comfyanonymous 2025-03-09 03:30:20 -04:00
  • 7395b0c0d1 Support new hunyuan video i2v model. comfyanonymous 2025-03-08 20:25:14 -05:00
  • 0952569493 Fix stable cascade VAE on some lowvram machines. comfyanonymous 2025-03-08 20:24:04 -05:00
  • 29832b3b61 Warn if frontend package is older than the one in requirements.txt comfyanonymous 2025-03-08 03:51:36 -05:00
  • be4e760648 Add an image_interleave option to the Hunyuan image to video encode node. comfyanonymous 2025-03-07 19:56:11 -05:00
  • c3d9cc4592 Print the frontend version in the log. comfyanonymous 2025-03-07 19:53:07 -05:00
  • 84cc9cb528
    Update frontend to 1.11.8 (#7119) Chenlei Hu 2025-03-07 19:02:13 -05:00
  • ebbb920163 Add back taesd to nightly package. comfyanonymous 2025-03-07 14:56:09 -05:00
  • d60fe0af4a Reduce size of nightly package. comfyanonymous 2025-03-07 08:30:01 -05:00
  • 5dbd250965 Update nightly instructions in readme. comfyanonymous 2025-03-07 07:57:59 -05:00
  • 4ab1875283 Add .bat file to nightly package to run with fp16 accumulation. comfyanonymous 2025-03-07 07:45:40 -05:00
  • 11b1f27cb1 Set WAN default compute dtype to fp16. comfyanonymous 2025-03-07 04:52:36 -05:00
  • 70e15fd743 No need for scale_input when fp8 matrix mult is disabled. comfyanonymous 2025-03-07 04:49:20 -05:00
  • e1474150de Support fp8_scaled diffusion models that don't use fp8 matrix mult. comfyanonymous 2025-03-07 04:37:58 -05:00
  • e62d72e8ca
    Typo in node_typing.py (#7092) JettHu 2025-03-07 04:24:04 +08:00
  • 1650cda030
    Fixed: Incorrect guide message for missing frontend. (#7105) Dr.Lt.Data 2025-03-07 05:23:23 +09:00
  • a13125840c ComfyUI version v0.3.24 comfyanonymous 2025-03-06 13:53:48 -05:00
  • dfa36e6855 Fix some things breaking when embeddings fail to apply. comfyanonymous 2025-03-06 13:31:40 -05:00
  • 0124be4d93 ComfyUI version v0.3.23 comfyanonymous 2025-03-06 04:10:12 -05:00
  • 29a70ca101 Support HunyuanVideo image to video model. comfyanonymous 2025-03-06 03:07:15 -05:00
  • 0bef826a98 Support llava clip vision model. comfyanonymous 2025-03-06 00:24:43 -05:00
  • 85ef295069 Make applying embeddings more efficient. comfyanonymous 2025-03-05 17:34:38 -05:00
  • 5d84607bf3
    Add type hint for FileLocator (#6968) Chenlei Hu 2025-03-05 15:35:26 -05:00
  • c1909f350f
    Better argument handling of front-end-root (#7043) Silver 2025-03-05 21:34:22 +01:00
  • 52b3469606
    [NodeDef] Explicitly add control_after_generate to seed/noise_seed (#7059) Chenlei Hu 2025-03-05 15:33:23 -05:00
  • 889519971f Bump ComfyUI version to v0.3.22 comfyanonymous 2025-03-05 10:06:37 -05:00
  • 76739c23c3 Revert "Partially revert last commit." comfyanonymous 2025-03-05 09:57:40 -05:00
  • a80bc822a2 Partially revert last commit. comfyanonymous 2025-03-05 08:58:44 -05:00
  • 872780d236
    fix: ltxv crop guides works with 0 keyframes (#7085) Andrew Kvochko 2025-03-05 15:47:32 +02:00
  • 6d45ffbe23 Bump ComfyUI version to v0.3.21 comfyanonymous 2025-03-05 08:05:22 -05:00
  • 77633ba77d Remove unused variable. comfyanonymous 2025-03-05 07:31:47 -05:00
  • 30e6cfb1a0 Fix LTXVPreprocess on resolutions that are not multiples of 2. comfyanonymous 2025-03-05 07:18:13 -05:00
  • dc134b2fdb Bump ComfyUI version to v0.3.20 comfyanonymous 2025-03-05 06:28:14 -05:00
  • 369b079ff6 Fix lowvram issue with ltxv vae. comfyanonymous 2025-03-05 05:26:08 -05:00
  • 9c9a7f012a Adjust ltxv memory factor. comfyanonymous 2025-03-05 05:16:05 -05:00
  • 93fedd92fe Support LTXV 0.9.5. comfyanonymous 2025-03-05 00:13:49 -05:00
  • 745b13649b Add update instructions for the portable. comfyanonymous 2025-03-04 23:34:36 -05:00
  • 2b140654c7
    suggest absolute full path to the requirements.txt instead of just requirements.txt (#7079) Dr.Lt.Data 2025-03-05 13:29:34 +09:00
  • 65042f7d39 Make it easier to set a custom template for hunyuan video. comfyanonymous 2025-03-04 09:26:05 -05:00
  • 7c7c70c400 Refactor skyreels i2v code. comfyanonymous 2025-03-04 00:15:45 -05:00
  • 8362199ee7 Bump ComfyUI version to v0.3.19 comfyanonymous 2025-03-03 19:18:37 -05:00
  • f86c724ef2 Temporal area composition. comfyanonymous 2025-03-03 06:50:31 -05:00
  • d6e5d487ad
    improved: better frontend package installation guide (#7047) Dr.Lt.Data 2025-03-03 18:40:23 +09:00
  • 6752a826f6 Make the missing frontend package error more obvious. comfyanonymous 2025-03-02 15:43:56 -05:00
  • 04cf0ccb51
    Use comfyui_frontend_package pypi package to manage frontend dependency (Frontend v1.10.17) (#7021) Chenlei Hu 2025-03-02 14:18:33 -05:00
  • 9af6320ec9 Make 2d area composition nodes work on video models. comfyanonymous 2025-03-02 08:19:16 -05:00
  • 6f81cd8973 Change defaults in WanImageToVideo node. comfyanonymous 2025-03-01 19:26:48 -05:00
  • 4dc6709307 Rename argument in last commit and document the options. comfyanonymous 2025-03-01 02:43:49 -05:00
  • 4d55f16ae8
    Use enum list for --fast options (#7024) Chenlei Hu 2025-03-01 02:37:35 -05:00
  • cf0b549d48 --fast now takes a number as argument to indicate how fast you want it. comfyanonymous 2025-02-28 02:48:20 -05:00
  • eb4543474b Use fp16 for intermediate for fp8 weights with --fast if supported. comfyanonymous 2025-02-28 02:17:50 -05:00
  • 1804397952 Use fp16 if checkpoint weights are fp16 and the model supports it. comfyanonymous 2025-02-27 16:39:57 -05:00
  • f4dac8ab6f Wan code small cleanup. comfyanonymous 2025-02-27 07:22:42 -05:00
  • b07f116dea Bump ComfyUI version to v0.3.18 comfyanonymous 2025-02-26 21:19:14 -05:00
  • 714f728820 Add to README that the Wan model is supported. comfyanonymous 2025-02-26 20:48:50 -05:00
  • 92d8d15300 Readme changes. comfyanonymous 2025-02-26 20:47:08 -05:00
  • 89253e9fe5
    Support Cambricon MLU (#6964) BiologicalExplosion 2025-02-27 09:45:13 +08:00
  • 3ea3bc8546 Fix wan issues when prompt length is long. comfyanonymous 2025-02-26 20:34:02 -05:00
  • 8e69e2ddfd Bump ComfyUI version to v0.3.17 comfyanonymous 2025-02-26 17:59:10 -05:00
  • 0270a0b41c Reduce artifacts on Wan by doing the patch embedding in fp32. comfyanonymous 2025-02-26 16:59:26 -05:00
  • 26c7baf789 Bump ComfyUI version to v0.3.16 comfyanonymous 2025-02-26 14:30:32 -05:00
  • c37f15f98e Add fast preview support for Wan models. comfyanonymous 2025-02-26 08:56:23 -05:00
  • 4bca7367f3 Don't try to use clip_fea on t2v model. comfyanonymous 2025-02-26 08:38:09 -05:00
  • b6fefe686b Better wan memory estimation. comfyanonymous 2025-02-26 07:51:22 -05:00
  • fa62287f1f More code reuse in wan. comfyanonymous 2025-02-26 05:22:29 -05:00
  • 0844998db3 Slightly better wan i2v mask implementation. comfyanonymous 2025-02-26 03:49:50 -05:00
  • 4ced06b879 WIP support for Wan I2V model. comfyanonymous 2025-02-26 01:49:43 -05:00
  • cb06e9669b Wan seems to work with fp16. comfyanonymous 2025-02-25 21:37:12 -05:00
  • 0c32f82298 Fix missing frames in SaveWEBM node. comfyanonymous 2025-02-25 20:21:03 -05:00
  • 189da3726d
    Update README.md (#6960) Yoland Yan 2025-02-25 17:17:18 -08:00
  • 9a66bb972d Make wan work with all latent resolutions. comfyanonymous 2025-02-25 19:56:04 -05:00
  • ea0f939df3 Fix issue with wan and other attention implementations. comfyanonymous 2025-02-25 19:13:39 -05:00
  • f37551c1d2 Change wan rope implementation to the flux one. comfyanonymous 2025-02-25 19:11:14 -05:00
  • 63023011b9 WIP support for Wan t2v model. comfyanonymous 2025-02-25 17:20:35 -05:00
  • f40076096e Cleanup some lumina te code. comfyanonymous 2025-02-25 04:10:26 -05:00
  • 96d891cb94 Speedup on some models by not upcasting bfloat16 to float32 on mac. comfyanonymous 2025-02-24 05:41:07 -05:00
  • 4553891bbd
    Update installation documentation to include desktop + cli. (#6899) Robin Huang 2025-02-23 16:13:39 -08:00
  • ace899e71a Prioritize fp16 compute when using allow_fp16_accumulation comfyanonymous 2025-02-23 04:45:54 -05:00
  • aff16532d4 Remove some useless code. comfyanonymous 2025-02-22 04:45:14 -05:00
  • b50ab153f9 Bump ComfyUI version to v0.3.15 comfyanonymous 2025-02-21 20:28:28 -05:00
  • 072db3bea6 Assume the mac black image bug won't be fixed before v16. comfyanonymous 2025-02-21 20:24:07 -05:00
  • a6deca6d9a Latest mac still has the black image bug. comfyanonymous 2025-02-21 20:14:30 -05:00
  • 41c30e92e7 Let all model memory be offloaded on nvidia. comfyanonymous 2025-02-21 06:32:11 -05:00
  • f579a740dd
    Update frontend release schedule in README. (#6908) filtered 2025-02-21 21:58:12 +11:00
  • d37272532c
    Add discord channel to support section. (#6900) Robin Huang 2025-02-20 15:26:16 -08:00
  • 12da6ef581 Apparently directml supports fp16. comfyanonymous 2025-02-20 09:29:59 -05:00
  • 29d4384a75
    Normalize extra_model_config.yaml paths to prevent duplicates. (#6885) Robin Huang 2025-02-20 04:09:45 -08:00
  • c5be423d6b
    Fix link pointing to non-exisiting docs (#6891) Silver 2025-02-20 13:07:07 +01:00
  • b4d3652d88
    fixed: crash caused by outdated incompatible aiohttp dependency (#6841) Dr.Lt.Data 2025-02-19 21:15:36 +09:00