Commit Graph

  • 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
  • c441048a4f Make VAE Encode tiled node work with video VAE. comfyanonymous 2024-12-19 05:31:39 -05:00
  • 9f4b181ab3 Add fast previews for hunyuan video. v0.3.8 comfyanonymous 2024-12-18 18:24:23 -05:00
  • cbbf077593 Small optimizations. comfyanonymous 2024-12-18 18:23:28 -05:00
  • 0c04a6ae78 Add .github folder to maintainer owner list (#6027) Chenlei Hu 2024-12-18 12:06:53 -08:00
  • 416ccc9e45 Update web content to release v1.5.19 (#6105) Chenlei Hu 2024-12-18 12:06:20 -08:00
  • ff2ff02168 Support old diffusion-pipe hunyuan video loras. comfyanonymous 2024-12-18 06:21:29 -05:00
  • 4c5c4ddeda Fix regression in VAE code on old pytorch versions. comfyanonymous 2024-12-18 03:08:28 -05:00
  • 79badea452 Add ConditioningStableAudio. comfyanonymous 2024-12-18 03:01:12 -05:00
  • 37e5390f5f Add: --use-sage-attention to enable SageAttention. comfyanonymous 2024-12-18 01:56:10 -05:00
  • a4f59bc65e Pick attention implementation based on device in llama code. comfyanonymous 2024-12-18 01:30:20 -05:00
  • ca457f7ba1 Properly tokenize the template for hunyuan video. comfyanonymous 2024-12-17 16:18:35 -05:00
  • cd6f615038 Fix tiled vae not working with some shapes. comfyanonymous 2024-12-17 09:42:17 -05:00
  • 517669aaa3 add preview 3d node (#6070) Terry Jia 2024-12-17 13:42:24 -05:00
  • e4e1bff605 Support diffusion-pipe hunyuan video lora format. comfyanonymous 2024-12-17 07:14:21 -05:00
  • d6656b0c0c Support llama hunyuan video text encoder in scaled fp8 format. comfyanonymous 2024-12-17 04:19:22 -05:00
  • f4cdedea62 Fix regression with ltxv VAE. comfyanonymous 2024-12-17 02:17:31 -05:00
  • 39b1fc4ccc Adjust used dtypes for hunyuan video VAE and diffusion model. comfyanonymous 2024-12-16 23:31:10 -05:00
  • 0b25f47bd9 Add some missing imports. comfyanonymous 2024-12-16 19:42:01 -05:00
  • bda1482a27 Basic Hunyuan Video model support. comfyanonymous 2024-12-16 19:35:40 -05:00
  • 19ee5d9d8b Don't expand mask when not necessary. comfyanonymous 2024-12-16 18:22:50 -05:00
  • 61b50720d0 Add support for attention masking in Flux (#5942) Raphael Walker 2024-12-17 00:21:17 +01:00
  • 0f954f34af Update README.md (#6071) Alexander Dyadyun 2024-12-17 03:24:54 +07:00
  • 5262901c5c Update web content to release v1.5.18 (#6075) Chenlei Hu 2024-12-16 11:38:24 -08:00
  • cc550d5908 use String directly to set bg color for load 3d canvas (#6057) Terry Jia 2024-12-16 13:51:40 -05:00
  • 6d1a3f7d00 Fix case of ExecutionBlocker not handled correctly with INPUT_IS_LIST. comfyanonymous 2024-12-15 08:41:35 -05:00
  • 1b3a650f19 (fix): added "model_type" to photomaker node (#6047) Alexander Piskun 2024-12-15 08:18:02 +03:00
  • e83063bf24 Support conv3d in PatchEmbed. comfyanonymous 2024-12-14 05:29:30 -05:00
  • 558b7d8b22 fix: prestartup script is not applied due to extra_model_paths.yaml and ensure custom paths are used during startup (#5872) Dr.Lt.Data 2024-12-14 08:21:32 +09:00
  • caf2074773 add_model_folder_path: ensure unique paths by removing duplicates (#5998) Alexander Piskun 2024-12-14 02:19:22 +03:00
  • bdf393792d add load 3d node support (#5564) Terry Jia 2024-12-13 18:13:52 -05:00
  • 4e14032c02 Make pad_to_patch_size function work on multi dim. comfyanonymous 2024-12-13 07:21:52 -05:00
  • 59d58b1158 [Security] Fix potential XSS on /view (#6034) Chenlei Hu 2024-12-13 01:56:43 -08:00
  • 563291ee51 Enforce all pyflake lint rules (#6033) Chenlei Hu 2024-12-12 16:29:37 -08:00
  • 6c0377f43e Enforce F821 undefined-name (#6032) Chenlei Hu 2024-12-12 16:24:41 -08:00
  • 2cddbf0821 Lint and fix undefined names (1/N) (#6028) Chenlei Hu 2024-12-12 15:55:26 -08:00
  • 60749f345d Lint and fix undefined names (3/N) (#6030) Chenlei Hu 2024-12-12 15:49:40 -08:00
  • d4426dce7c Lint and fix undefined names (2/N) (#6029) Chenlei Hu 2024-12-12 15:48:21 -08:00
  • d9d7f3c619 Lint all unused variables (#5989) Chenlei Hu 2024-12-12 14:59:16 -08:00
  • fd5dfb812c Set initial load devices for te and model to mps device on mac. comfyanonymous 2024-12-12 06:00:31 -05:00
  • 3dfdddcc91 Update README (Add new keybinding entries) (#6020) Chenlei Hu 2024-12-11 15:55:38 -08:00
  • 5747bc6457 Optimize model library (#5841) Hayden 2024-12-12 07:12:04 +08:00
  • 5bea1d2ec9 Add MaHiRo (improved/alternate CFG) (#5975) yoinked 2024-12-11 13:51:51 -08:00
  • 5def9fbc83 Update CI workflow to remove Windows testing configuration (#6007) Yoland Yan 2024-12-11 13:48:41 -08:00
  • 7a7efe8424 Support loading some checkpoint files with nested dicts. comfyanonymous 2024-12-11 08:04:54 -05:00
  • 44db978531 Fix a few things in text enc code for models with no eos token. comfyanonymous 2024-12-10 23:07:26 -05:00
  • 1c8d11e48a Support different types of tokenizers. comfyanonymous 2024-12-10 09:44:13 -05:00
  • a220d11e6b Replace pylint with ruff (#5987) Chenlei Hu 2024-12-09 19:04:23 -08:00
  • 23827ca312 Add cond_scale to sampler_post_cfg_function (#5985) catboxanon 2024-12-09 20:13:18 -05:00
  • 0fd4e6c778 Lint unused import (#5973) Chenlei Hu 2024-12-09 14:24:39 -06:00
  • e2fafe0686 Make CLIP set last layer node work with t5 models. comfyanonymous 2024-12-09 03:57:14 -05:00
  • 6579632201 Remove unused imports and variables. comfyanonymous 2024-12-08 08:06:00 -05:00
  • ac2f0523ca Set env vars to disable telemetry in libs used by some custom nodes. comfyanonymous 2024-12-07 14:51:20 -05:00
  • fbf68c4e52 clamp input (#5928) Haoming 2024-12-08 03:00:31 +08:00
  • 93477f8efe Add code owners (#5873) Chenlei Hu 2024-12-06 19:00:54 -08:00
  • 8af9a91e0c A few improvements to #5937. comfyanonymous 2024-12-06 05:48:34 -05:00
  • 005d2d3a13 ltxv: add noise to guidance image to ensure generated motion. (#5937) Michael Kupchick 2024-12-06 12:46:08 +02:00
  • 1e21f4c14e Make timestep ranges more usable on rectified flow models. comfyanonymous 2024-12-05 16:40:58 -05:00
  • 9a616b81c1 Add rescaling_scale from STG to SkipLayerGuidanceDiT. v0.3.7 comfyanonymous 2024-12-04 19:25:50 -05:00
  • 3bed56bb13 Add another ROCm tip. comfyanonymous 2024-12-04 07:11:16 -05:00
  • 4e402b11c6 Reland union type (#5900) filtered 2024-12-05 07:12:10 +11:00
  • 48272448ad [Developer Experience] Add node typing (#5676) Chenlei Hu 2024-12-04 12:01:00 -08:00
  • f7695b5f9e Add Create Hook Keyframes Interp. node to simplify creating groups of keyframes without external nodes (#5896) Jedrzej Kosinski 2024-12-03 20:03:09 -06:00
  • 452179fe4f Make ModelPatcher class clone function work with inheritance. comfyanonymous 2024-12-03 13:57:18 -05:00
  • bf9a90a145 Revert "Add union link connection type support (#5806)" (#5889) Chenlei Hu 2024-12-03 10:06:34 -08:00
  • c1b92b719d Some optimizations to euler a. comfyanonymous 2024-12-03 06:11:52 -05:00
  • cdc3b97dd5 resolve relative paths in YAML configuration for extra model paths (#5847) Alexander Piskun 2024-12-03 14:02:01 +03:00
  • 8d4e06324f Add union link connection type support (#5806) Chenlei Hu 2024-12-03 02:46:00 -08:00
  • 57e8bf6a9f Fix case where a memory leak could cause crash. comfyanonymous 2024-12-02 19:49:49 -05:00
  • 0ee322ec5f ModelPatcher Overhaul and Hook Support (#5583) Jedrzej Kosinski 2024-12-02 13:51:02 -06:00
  • 79d5ceae6e Improved memory management. (#5450) comfyanonymous 2024-12-02 14:39:34 -05:00
  • 2d5b3e0078 Remove useless code. comfyanonymous 2024-12-02 06:49:55 -05:00
  • 8e4118c0de make dpm_2_ancestral work with rectified flow. v0.3.6 comfyanonymous 2024-12-01 07:37:41 -05:00
  • 3fc6ebcdd7 Add basic style model "multiply" strength. comfyanonymous 2024-11-30 07:27:11 -05:00
  • 20a560eb97 How to enable experimental memory efficient attention on ROCm RDNA3. comfyanonymous 2024-11-29 06:18:56 -05:00
  • 82c5308561 Backward compatibility patch for changes in the method signature of InpaintModelConditioning. (#5825) Dr.Lt.Data 2024-11-29 10:30:28 +09:00
  • 26fb2c68e8 Add a way to disable cropping in the CLIPVisionEncode node. comfyanonymous 2024-11-28 20:24:47 -05:00
  • bf2650a80e Fast previews for ltxv. comfyanonymous 2024-11-28 06:46:15 -05:00
  • 53646e0f32 Update web content to release v1.4.13 (#5807) Chenlei Hu 2024-11-28 01:59:06 -08:00
  • 20879c78f9 Remove internal model download endpoint (#5432) Chenlei Hu 2024-11-28 01:57:06 -08:00
  • b666539595 Remove print. comfyanonymous 2024-11-27 20:28:39 -05:00
  • 95d8713482 Missing parentheses. comfyanonymous 2024-11-27 13:45:32 -05:00
  • 0d4e29f13f LTXV model merging node. comfyanonymous 2024-11-27 01:42:00 -05:00
  • 497db6212f Alternative fix for #5767 comfyanonymous 2024-11-26 17:51:40 -05:00
  • 24dc581dc3 fix multi add makedirs error (#5786) lky 2024-11-27 04:34:19 +08:00
  • 4c82741b54 Support official SD3.5 Controlnets. v0.3.5 comfyanonymous 2024-11-26 11:31:25 -05:00
  • 15c39ea757 Support for the official mochi lora format. comfyanonymous 2024-11-26 03:34:36 -05:00
  • b7143b74ce Flux inpaint model does not work in fp16. comfyanonymous 2024-11-26 01:33:01 -05:00
  • 61196d8857 Add option to inference the diffusion model in fp32 and fp64. comfyanonymous 2024-11-25 05:00:23 -05:00
  • b4526d3fc3 Skip layer guidance now works on hydit model. comfyanonymous 2024-11-24 05:54:30 -05:00
  • 3d802710e7 Update README.md (#5707) 40476 2024-11-24 04:12:07 -05:00
  • 7126ecffde set LTX min length to 1 for t2i (#5750) spacepxl 2024-11-23 21:33:08 -05:00
  • ab885b33ba Skip layer guidance node now works on LTX-Video. comfyanonymous 2024-11-23 10:33:05 -05:00
  • 839ed3368e Some improvements to the lowvram unloading. comfyanonymous 2024-11-22 20:59:15 -05:00
  • 6e8cdcd3cb Fix some tiled VAE decoding issues with LTX-Video. v0.3.4 comfyanonymous 2024-11-22 18:00:34 -05:00
  • e5c3f4b87f LTXV lowvram fixes. comfyanonymous 2024-11-22 17:17:11 -05:00
  • bc6be6c11e Some fixes to the lowvram system. v0.3.3 comfyanonymous 2024-11-22 16:40:04 -05:00