wuyingwen/comfyui
1
0
Fork 0
Code Issues Pull Requests Actions 5 Packages Projects Releases Wiki Activity

Commit Graph

  • 6320d05696 Slightly lower hunyuan video memory usage. comfyanonymous 2025-01-16 00:23:01 -05:00
  • 25683b5b02 Lower cosmos diffusion model memory usage. comfyanonymous 2025-01-15 23:46:42 -05:00
  • 4758fb64b9 Lower cosmos VAE memory usage by a bit. comfyanonymous 2025-01-15 22:57:52 -05:00
  • 008761166f Optimize first attention block in cosmos VAE. comfyanonymous 2025-01-15 21:48:46 -05:00
  • bfd5dfd611 3.13 doesn't work yet. comfyanonymous 2025-01-15 20:32:44 -05:00
  • 55ade36d01 Remove python 3.8 from test-build workflow. comfyanonymous 2025-01-15 20:24:55 -05:00
  • 2e20e399ea Add minimum numpy version to requirements.txt comfyanonymous 2025-01-15 20:19:56 -05:00
  • 3baf92d120 CosmosImageToVideoLatent batch_size now does something. comfyanonymous 2025-01-15 17:19:59 -05:00
  • 1709a8441e Use latest python 3.12.8 the portable release. comfyanonymous 2025-01-15 14:50:40 -05:00
  • cba58fff0b Remove unsafe embedding load for very old pytorch. comfyanonymous 2025-01-15 04:32:23 -05:00
  • 2feb8d0b77 Force safe loading of files in torch format on pytorch 2.4+ comfyanonymous 2025-01-15 03:50:27 -05:00
  • 5b657f8c15 Allow setting start and end image in CosmosImageToVideoLatent. comfyanonymous 2025-01-15 00:41:35 -05:00
  • 2cdbaf5169
         Add SetFirstSigma node (#6459) catboxanon 2025-01-14 19:05:45 -05:00
  • c78a45685d
         Rewrite res_multistep sampler and implement res_multistep_cfg_pp sampler. (#6462) Pam 2025-01-15 04:20:06 +05:00
  • 3aaabb12d4 Implement Cosmos Image/Video to World (Video) diffusion models. comfyanonymous 2025-01-14 05:14:10 -05:00
  • 1f1c7b7b56 Remove useless code. comfyanonymous 2025-01-13 03:52:37 -05:00
  • 90f349f93d Add res_multistep sampler from the cosmos code. comfyanonymous 2025-01-12 03:09:20 -05:00
  • b9d9bcba14
         fixed a bug where a relative path was not converted to a full path (#6395) Alexander Piskun 2025-01-12 03:19:51 +03:00
  • 42086af123
         Merge ruff.toml into pyproject.toml (#6431) Chenlei Hu 2025-01-11 12:52:46 -05:00
  • 6c9bd11fa3
         Hooks Part 2 - TransformerOptionsHook and AdditionalModelsHook (#6377) Jedrzej Kosinski 2025-01-11 11:20:23 -06:00
  • ee8a7ab69d Fast latent preview for Cosmos. comfyanonymous 2025-01-11 04:40:45 -05:00
  • 9c773a241b
         Add pyproject.toml (#6386) Chenlei Hu 2025-01-11 03:09:25 -05:00
  • adea2beb5c Add edm option to ModelSamplingContinuousEDM for Cosmos. comfyanonymous 2025-01-11 02:18:42 -05:00
  • 2ff3104f70 WIP support for Nvidia Cosmos 7B and 14B text to world (video) models. comfyanonymous 2025-01-10 09:11:57 -05:00
  • 129d8908f7 Add argument to skip the output reshaping in the attention functions. comfyanonymous 2025-01-10 06:27:37 -05:00
  • ff838657fa Cleaner handling of attention mask in ltxv model code. comfyanonymous 2025-01-09 07:12:03 -05:00
  • 2307ff6746 Improve some logging messages. comfyanonymous 2025-01-08 19:05:22 -05:00
  • d0f3752e33 Properly calculate inner dim for t5 model. comfyanonymous 2025-01-07 17:32:29 -05:00
  • c515bdf371
         fixed: robust loading comfy.settings.json (#6383) Dr.Lt.Data 2025-01-08 06:03:56 +09:00
  • 4209edf48d Make a few more samplers deterministic. comfyanonymous 2025-01-07 02:12:12 -05:00
  • d055325783
         Document get_attr and get_model_object (#6357) Chenlei Hu 2025-01-06 20:12:22 -05:00
  • eeab420c70
         Update frontend to v1.6.18 (#6368) Chenlei Hu 2025-01-06 18:42:45 -05:00
  • 916d1e14a9 Make ancestral samplers more deterministic. comfyanonymous 2025-01-06 03:01:04 -05:00
  • c496e53519
         In inner_sample, change "sigmas" to "sampler_sigmas" in transformer_options to not conflict with the "sigmas" that will overwrite "sigmas" in _calc_cond_batch (#6360) Jedrzej Kosinski 2025-01-06 00:36:47 -06:00
  • 7da85fac3f
         Update CODEOWNERS (#6338) Yoland Yan 2025-01-05 01:33:49 -08:00
  • b65b83af6f
         Add update-frontend github action (#6336) Chenlei Hu 2025-01-05 04:32:11 -05:00
  • c8a3492c22 Make the device an optional parameter in the clip loaders. comfyanonymous 2025-01-05 04:29:36 -05:00
  • 5cbf79787f Add advanced device option to clip loader nodes. comfyanonymous 2025-01-05 01:46:11 -05:00
  • d45ebb63f6 Remove old unused function. comfyanonymous 2025-01-04 07:20:54 -05:00
  • caa6476a69
         Update web content to release v1.6.17 (#6337) Chenlei Hu 2025-01-03 16:22:08 -05:00
  • 45671cda0b
         Update web content to release v1.6.16 (#6335) Chenlei Hu 2025-01-03 13:56:46 -05:00
  • 8f29664057 Change defaults in nightly package workflow. comfyanonymous 2025-01-03 12:06:01 -05:00
  • 0b9839ef43
         Update web content to release v1.6.15 (#6324) Chenlei Hu 2025-01-02 19:20:48 -05:00
  • 953693b137
         add fov and mask for load 3d node (#6308) Terry Jia 2025-01-02 19:20:34 -05:00
  • a39ea87bca
         Update web content to release v1.6.14 (#6312) Chenlei Hu 2025-01-02 16:18:54 -05:00
  • 9e9c8a1c64 Clear cache as often on AMD as Nvidia. comfyanonymous 2025-01-02 08:44:16 -05:00
  • 0f11d60afb
         Fix temporal tiling for decoder, remove redundant tiles. (#6306) Andrew Kvochko 2025-01-01 21:29:01 +00:00
  • 79eea51a1d Fix and enforce all ruff W rules. comfyanonymous 2025-01-01 03:08:33 -05:00
  • c0338a46a4
         Fix unknown sampler error handling in calculate_sigmas function (#6280) blepping 2024-12-31 15:33:50 -07:00
  • 1c99734e5a
         Add missing model_options param (#6296) Jedrzej Kosinski 2024-12-31 13:46:55 -06:00
  • 67758f50f3
         Fix custom node type-hinting examples (#6281) filtered 2024-12-31 19:41:09 +11:00
  • 02eef72bf5
         fixed "verbose" argument (#6289) Alexander Piskun 2024-12-31 11:27:09 +03:00
  • b7572b2f87 Fix and enforce no trailing whitespace. comfyanonymous 2024-12-31 03:16:37 -05:00
  • a90aafafc1
         Add kl_optimal scheduler (#6206) blepping 2024-12-30 03:09:38 -07:00
  • d9b7cfac7e Fix and enforce new lines at the end of files. comfyanonymous 2024-12-30 04:14:59 -05:00
  • 3507870535
         Add 'sigmas' to transformer_options so that downstream code can know about the full scope of current sampling run, fix Hook Keyframes' guarantee_steps=1 inconsistent behavior with sampling split across different Sampling nodes/sampling runs by referencing 'sigmas' (#6273) Jedrzej Kosinski 2024-12-30 02:42:49 -06:00
  • 82ecb02c1e
         Remove duplicate calls to INPUT_TYPES (#6249) catboxanon 2024-12-29 20:06:49 -05:00
  • a618f768e0 Auto reshape 2d to 3d latent for single image generation on video model. comfyanonymous 2024-12-29 02:26:49 -05:00
  • e1dec3c792 Fix formatting. comfyanonymous 2024-12-28 05:33:17 -05:00
  • 96697c4bc5
         serve workflow templates from custom_nodes (#6193) Zoltán Dócs 2024-12-28 11:30:04 +01:00
  • b504bd606d Add ruff rule for empty line with trailing whitespace. comfyanonymous 2024-12-28 05:22:21 -05:00
  • d170292594 Remove some trailing white space. comfyanonymous 2024-12-27 18:02:21 -05:00
  • 9cfd185676
         Add option to log non-error output to stdout (#6243) filtered 2024-12-28 06:40:05 +11:00
  • 4b5bcd8ac4 Closer memory estimation for hunyuan dit model. comfyanonymous 2024-12-27 07:37:00 -05:00
  • ceb50b2cbf Closer memory estimation for pixart models. comfyanonymous 2024-12-27 07:30:09 -05:00
  • 160ca08138 Use python 3.9 in launch test instead of 3.8 comfyanonymous 2024-12-26 20:05:54 -05:00
  • c4bfdba330
         Support ascend npu (#5436) Huazhong Ji 2024-12-27 08:36:50 +08:00
  • ee9547ba31 Improve temporal VAE Encode (Tiled) math. comfyanonymous 2024-12-26 07:18:49 -05:00
  • 19a64d6291 Cleanup some mac related code. comfyanonymous 2024-12-25 05:32:51 -05:00
  • b486885e08 Disable bfloat16 on older mac. comfyanonymous 2024-12-25 05:18:50 -05:00
  • 0229228f3f Clean up the VAE dtypes code. comfyanonymous 2024-12-25 04:50:34 -05:00
  • 1ed75ab30e Update nightly pytorch instructions in readme for nvidia. comfyanonymous 2024-12-25 03:29:03 -05:00
  • 99a1fb6027 Make fast fp8 take a bit less peak memory. comfyanonymous 2024-12-24 18:05:19 -05:00
  • 73e04987f7 Prevent black images in VAE Decode (Tiled) node. comfyanonymous 2024-12-24 07:36:30 -05:00
  • 5388df784a Add temporal tiling to VAE Encode (Tiled) node. comfyanonymous 2024-12-24 07:10:09 -05:00
  • 26e0ba8f8c
         Enable External Event Loop Integration for ComfyUI [refactor] (#6114) Alexander Piskun 2024-12-24 14:38:52 +03:00
  • bc6dac4327 Add temporal tiling to VAE Decode (Tiled) node. comfyanonymous 2024-12-23 20:03:37 -05:00
  • f18ebbd316
         Use raw dir name to serve static web content (#6107) Chenlei Hu 2024-12-23 03:29:42 -05:00
  • 15564688ed Add a try except block so if torch version is weird it won't crash. comfyanonymous 2024-12-23 03:22:48 -05:00
  • c6b9c11ef6
         Add oneAPI device selector for xpu and some other changes. (#6112) Simon Lui 2024-12-23 00:18:32 -08:00
  • e44d0ac7f7 Make --novram completely offload weights. comfyanonymous 2024-12-23 01:50:11 -05:00
  • 56bc64f351 Comment out some useless code. comfyanonymous 2024-12-22 23:51:14 -05:00
  • f7d83b72e0
         fixed a bug in ldm/pixart/blocks.py (#6158) zhangp365 2024-12-23 12:44:20 +08:00
  • 80f07952d2 Fix lowvram issue with ltxv vae. comfyanonymous 2024-12-22 23:20:17 -05:00
  • 57f330caf9 Relax minimum ratio of weights loaded in memory on nvidia. comfyanonymous 2024-12-22 03:06:37 -05:00
  • 601ff9e3db Add that Hunyuan Video and Pixart are supported to readme. comfyanonymous 2024-12-21 11:31:39 -05:00
  • 341667c4d5
         remove minimum step count for AYS (#6137) TechnoByte 2024-12-21 16:05:09 +01:00
  • 1419dee915
         Update README.md for Intel GPUs (#6069) Qiacheng Li 2024-12-20 15:04:03 -08:00
  • da13b6b827 Get rid of meshgrid warning. comfyanonymous 2024-12-20 18:02:12 -05:00
  • c86cd58573 Remove useless code. comfyanonymous 2024-12-20 17:50:03 -05:00
  • b5fe39211a Remove some useless code. comfyanonymous 2024-12-20 17:43:50 -05:00
  • e946667216 Some fixes/cleanups to pixart code. comfyanonymous 2024-12-20 17:10:52 -05:00
  • d7969cb070
         Replace print with logging (#6138) Chenlei Hu 2024-12-20 13:24:55 -08:00
  • bddb02660c
         Add PixArt model support (#6055) City 2024-12-20 21:25:00 +01:00
  • 418eb7062d Support new LTXV VAE. comfyanonymous 2024-12-20 04:38:29 -05:00
  • cac68ca813 Fix some more video tiled encode issues. comfyanonymous 2024-12-19 23:14:03 -05:00
  • 52c1d933b2 Fix tiled hunyuan video VAE encode issue. comfyanonymous 2024-12-19 22:52:37 -05:00
  • 3cacd3fca5
         Support preview images embedded in safetensors metadata (#6119) catboxanon 2024-12-19 17:01:56 -05:00
  • 2dda7c11a3 More proper fix for the memory issue. comfyanonymous 2024-12-19 16:21:56 -05:00
  • 3ad3248ad7 Fix lowvram bug when using a model multiple times in a row. comfyanonymous 2024-12-19 16:04:56 -05:00