Commit Graph

  • bfce723311 Initial work on multigpu_clone function, which will account for additional_models getting cloned Jedrzej Kosinski 2025-01-17 03:31:28 -0600
  • 219237379c update clean uv use khaangnguyeen 2025-01-17 10:55:19 +0700
  • 7fc3ccdcc2 Add that nvidia cosmos is supported to the README. comfyanonymous 2025-01-16 21:17:18 -0500
  • 31f5458938 Merge branch 'master' into multigpu_support Jedrzej Kosinski 2025-01-16 18:25:05 -0600
  • ad7199bf39
    Merge f96fd4b00b into 55add50220 Navonil Majumder 2025-01-16 18:23:38 -0600
  • 55add50220 Bump ComfyUI version to v0.3.12 v0.3.12 comfyanonymous 2025-01-16 18:11:57 -0500
  • 0aa2368e46 Fix some cosmos fp8 issues. comfyanonymous 2025-01-16 17:45:37 -0500
  • cca96a85ae Fix cosmos VAE failing with videos longer than 121 frames. comfyanonymous 2025-01-16 16:30:06 -0500
  • 619b8cde74 Bump ComfyUI version to 0.3.11 v0.3.11 comfyanonymous 2025-01-16 14:54:48 -0500
  • 31831e6ef1 Code refactor. comfyanonymous 2025-01-16 07:23:54 -0500
  • 88ceb28e20 Tweak hunyuan memory usage factor. comfyanonymous 2025-01-16 06:31:03 -0500
  • 23289a6a5c Clean up some debug lines. comfyanonymous 2025-01-16 04:24:39 -0500
  • 9d8b6c1f46 More accurate memory estimation for cosmos and hunyuan video. comfyanonymous 2025-01-16 03:48:40 -0500
  • 6320d05696 Slightly lower hunyuan video memory usage. comfyanonymous 2025-01-16 00:23:01 -0500
  • 25683b5b02 Lower cosmos diffusion model memory usage. comfyanonymous 2025-01-15 23:46:42 -0500
  • 4758fb64b9 Lower cosmos VAE memory usage by a bit. comfyanonymous 2025-01-15 22:57:52 -0500
  • 008761166f Optimize first attention block in cosmos VAE. comfyanonymous 2025-01-15 21:48:46 -0500
  • 2145a202eb Merge branch 'master' into multigpu_support Jedrzej Kosinski 2025-01-15 19:58:28 -0600
  • bfd5dfd611 3.13 doesn't work yet. comfyanonymous 2025-01-15 20:32:44 -0500
  • 55ade36d01 Remove python 3.8 from test-build workflow. comfyanonymous 2025-01-15 20:24:55 -0500
  • 2e20e399ea Add minimum numpy version to requirements.txt comfyanonymous 2025-01-15 20:19:56 -0500
  • 3baf92d120 CosmosImageToVideoLatent batch_size now does something. comfyanonymous 2025-01-15 17:19:59 -0500
  • 1709a8441e Use latest python 3.12.8 the portable release. comfyanonymous 2025-01-15 14:50:40 -0500
  • cba58fff0b Remove unsafe embedding load for very old pytorch. comfyanonymous 2025-01-15 04:32:23 -0500
  • 2feb8d0b77 Force safe loading of files in torch format on pytorch 2.4+ comfyanonymous 2025-01-15 03:50:27 -0500
  • 1dfe30e275 ignore cache jcastle 2025-01-15 15:26:01 +0900
  • 34c04cfcf0 add user to git jcastle 2025-01-15 15:24:05 +0900
  • 5b657f8c15 Allow setting start and end image in CosmosImageToVideoLatent. comfyanonymous 2025-01-15 00:41:35 -0500
  • 2cdbaf5169
    Add SetFirstSigma node (#6459) catboxanon 2025-01-14 19:05:45 -0500
  • c78a45685d
    Rewrite res_multistep sampler and implement res_multistep_cfg_pp sampler. (#6462) Pam 2025-01-15 04:20:06 +0500
  • 25818dc848 Added a 'max_gpus' input Jedrzej Kosinski 2025-01-14 13:45:14 -0600
  • 198953cd08 Add nodes_multigpu.py to loaded nodes Jedrzej Kosinski 2025-01-14 12:24:55 -0600
  • 9f3e21673f
    Merge branch 'comfyanonymous:master' into res_multistep Pam 2025-01-14 18:28:17 +0500
  • 3aaabb12d4 Implement Cosmos Image/Video to World (Video) diffusion models. comfyanonymous 2025-01-14 05:14:10 -0500
  • ec16ee2f39 Merge branch 'master' into multigpu_support Jedrzej Kosinski 2025-01-13 20:21:06 -0600
  • d5088072fb Make test node for multigpu instead of storing it in just a local __init__.py Jedrzej Kosinski 2025-01-13 20:20:25 -0600
  • 9cac7f563d Rewrite res_multistep sampler and implement res_multistep_cfg_pp sampler. Pam 2025-01-14 05:12:03 +0500
  • d40e461a80 Add SetFirstSigma node catboxanon 2025-01-13 13:48:07 -0500
  • 6789f3d532 Allow FP16 accumulation with --fast catboxanon 2025-01-13 10:30:01 -0500
  • 1f1c7b7b56 Remove useless code. comfyanonymous 2025-01-13 03:52:37 -0500
  • 95e0d8862d
    Merge branch 'comfyanonymous:master' into main Gwangwon Jung 2025-01-13 09:20:06 +0900
  • 1a00fb968f
    Update extra_config.py dchatel 2025-01-12 14:47:01 +0100
  • 8c9b7bd00f
    Update extra_config.py dchatel 2025-01-12 14:09:38 +0100
  • 90f349f93d Add res_multistep sampler from the cosmos code. comfyanonymous 2025-01-12 03:09:20 -0500
  • f40ff25e67 remove some params from load 3d node Terry Jia 2025-01-11 23:30:52 -0500
  • 8d4b50158e Merge branch 'master' into multigpu_support Jedrzej Kosinski 2025-01-11 20:16:42 -0600
  • b9d9bcba14
    fixed a bug where a relative path was not converted to a full path (#6395) Alexander Piskun 2025-01-12 03:19:51 +0300
  • 92f831eb24 rebased onto "master" Alexander Piskun 2025-01-11 22:14:26 +0300
  • d83ab7d1d6 Merge branch 'refs/heads/master' into ruff/C408 Alexander Piskun 2025-01-11 22:03:50 +0300
  • 42086af123
    Merge ruff.toml into pyproject.toml (#6431) Chenlei Hu 2025-01-11 12:52:46 -0500
  • 6c9bd11fa3
    Hooks Part 2 - TransformerOptionsHook and AdditionalModelsHook (#6377) Jedrzej Kosinski 2025-01-11 11:20:23 -0600
  • f891576e48 Merge ruff.toml into pyproject.toml huchenlei 2025-01-11 11:36:22 -0500
  • ee8a7ab69d Fast latent preview for Cosmos. comfyanonymous 2025-01-11 04:40:45 -0500
  • 9c773a241b
    Add pyproject.toml (#6386) Chenlei Hu 2025-01-11 03:09:25 -0500
  • adea2beb5c Add edm option to ModelSamplingContinuousEDM for Cosmos. comfyanonymous 2025-01-11 02:18:42 -0500
  • e88c6c03ff Fix cond_cat to not try to cast anything that doesn't have a 'to' function Jedrzej Kosinski 2025-01-10 23:05:24 -0600
  • d3cf2b7b24
    Merge branch 'comfyanonymous:master' into multigpu_support Jedrzej Kosinski 2025-01-10 20:24:37 -0600
  • 79c526f5c0
    Only convert to unint8 if directml_enabled Bradley Reynolds 2025-01-10 16:12:59 -0600
  • 84cafe556f Rename version.py to comfyui_version.py huchenlei 2025-01-10 10:34:38 -0500
  • 2ff3104f70 WIP support for Nvidia Cosmos 7B and 14B text to world (video) models. comfyanonymous 2025-01-10 09:11:57 -0500
  • 129d8908f7 Add argument to skip the output reshaping in the attention functions. comfyanonymous 2025-01-10 06:27:37 -0500
  • f96fd4b00b README.md: Make a new section for audio + Add TangoFlux to the section Navonil Majumder 2025-01-10 17:41:17 +0800
  • ff838657fa Cleaner handling of attention mask in ltxv model code. comfyanonymous 2025-01-09 07:12:03 -0500
  • 8dc9d9e812
    Merge branch 'comfyanonymous:master' into main Gwangwon Jung 2025-01-09 18:54:35 +0900
  • 28d5474d76 uv init was removed. With a pyproject.toml file alrealdy in root folder, project is interpreted as ready from uv. Diogo Felipe de Melo Santiago 2025-01-08 23:59:19 -0300
  • e231676a77
    Merge branch 'comfyanonymous:master' into master Diogo Santiago 2025-01-08 23:57:03 -0300
  • 2307ff6746 Improve some logging messages. comfyanonymous 2025-01-08 19:05:22 -0500
  • 6c16e0f852 Don't run on fork PRs huchenlei 2025-01-08 18:01:34 -0500
  • 40485a8f91 nit huchenlei 2025-01-08 17:58:09 -0500
  • 9b7f463106 chore: Update version.py to match pyproject.toml github-actions 2025-01-08 22:56:39 +0000
  • 0e1164006b nit huchenlei 2025-01-08 17:56:17 -0500
  • 7562b89cfd Grant pr write permission huchenlei 2025-01-08 17:48:10 -0500
  • 18b2112e66 Fix commit huchenlei 2025-01-08 17:44:12 -0500
  • f0670923a1 Change trigger to PR huchenlei 2025-01-08 17:25:31 -0500
  • 698243cdc0 Add github action to sync version.py huchenlei 2025-01-08 17:20:23 -0500
  • 2f74b518b0 Static version file huchenlei 2025-01-08 15:49:40 -0500
  • bc684efaa4 doc huchenlei 2025-01-07 21:56:33 -0500
  • 8f1e33e7da Add pyproject.toml huchenlei 2025-01-07 16:56:37 -0500
  • 126cdc582b
    fixed a bug where a relative path was not converted to a full path bigcat88 2025-01-08 14:21:56 +0200
  • 7448f02b7c Initial proof of concept of giving splitting cond sampling between multiple GPUs Jedrzej Kosinski 2025-01-08 03:33:05 -0600
  • 871258aa72 Add get_all_torch_devices to get detected devices intended for current torch hardware device Jedrzej Kosinski 2025-01-07 21:06:03 -0600
  • 66838ebd39
    Merge branch 'comfyanonymous:master' into multigpu_support Jedrzej Kosinski 2025-01-07 20:11:27 -0600
  • 76b7c71b6e
    Use non-alias Sergii Dymchenko 2025-01-07 15:51:19 -0800
  • e4e2d82b0f
    Use torch.special.expm1 Sergii Dymchenko 2025-01-07 15:39:37 -0800
  • d0f3752e33 Properly calculate inner dim for t5 model. comfyanonymous 2025-01-07 17:32:29 -0500
  • b2b498d466 Change Save Image default value Andrés Zsögön 2025-01-07 19:11:43 -0300
  • c515bdf371
    fixed: robust loading comfy.settings.json (#6383) Dr.Lt.Data 2025-01-08 06:03:56 +0900
  • c8ed83ce09 fixed: robust loading comfy.settings.json Dr.Lt.Data 2025-01-07 23:03:35 +0900
  • 7333281698 Clean up a typehint Jedrzej Kosinski 2025-01-07 02:58:59 -0600
  • 3cd4c5cb0a Rename AddModelsHooks to AdditionalModelsHook, rename SetInjectionsHook to InjectionsHook (not yet implemented, but at least getting the naming figured out) Jedrzej Kosinski 2025-01-07 02:22:49 -0600
  • 4209edf48d Make a few more samplers deterministic. comfyanonymous 2025-01-07 02:12:12 -0500
  • 11c6d56037 Merge branch 'master' into hooks_part2 Jedrzej Kosinski 2025-01-07 01:01:53 -0600
  • 216fea15ee Made TransformerOptionsHook contribute to registered hooks properly, added some doc strings and removed a so-far unused variable Jedrzej Kosinski 2025-01-07 00:59:18 -0600
  • 58bf8815c8 Add a get_injections function to ModelPatcher Jedrzej Kosinski 2025-01-06 20:34:30 -0600
  • d055325783
    Document get_attr and get_model_object (#6357) Chenlei Hu 2025-01-06 20:12:22 -0500
  • fb281b3c52
    Update model_patcher.py Chenlei Hu 2025-01-06 19:08:50 -0500
  • a605201278
    Update model_patcher.py Chenlei Hu 2025-01-06 19:08:16 -0500
  • eeab420c70
    Update frontend to v1.6.18 (#6368) Chenlei Hu 2025-01-06 18:42:45 -0500
  • 1b38f5bf57 removed 4 whitespace lines to satisfy Ruff, Jedrzej Kosinski 2025-01-06 17:11:12 -0600
  • 2724ac4a60 Merge branch 'master' into hooks_part2 Jedrzej Kosinski 2025-01-06 17:04:24 -0600