Commit Graph

  • 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
  • 94323a26a7 Remove prints. v0.3.2 comfyanonymous 2024-11-22 10:51:31 -05:00
  • 5818f6cf51 Remove print. comfyanonymous 2024-11-22 10:49:15 -05:00
  • 0b734de449 Add LTX-Video support to the Readme. comfyanonymous 2024-11-22 09:24:03 -05:00
  • 5e16f1d24b Support Lightricks LTX-Video model. comfyanonymous 2024-11-22 08:44:42 -05:00
  • 2fd9c1308a Fix mask issue in some attention functions. comfyanonymous 2024-11-22 02:10:09 -05:00
  • 8f0009aad0 Support new flux model variants. v0.3.1 comfyanonymous 2024-11-21 08:38:23 -05:00
  • 41444b5236 Add some new weight patching functionality. comfyanonymous 2024-11-21 07:19:17 -05:00
  • 772e620e32 Update readme. comfyanonymous 2024-11-20 20:42:51 -05:00
  • 07f6eeaa13 Fix mask issue with attention_xformers. comfyanonymous 2024-11-20 17:07:46 -05:00
  • 22535d0589 Skip layer guidance now works on stable audio model. v0.3.0 comfyanonymous 2024-11-20 07:33:06 -05:00
  • 898615122f Rename add_noise_mask -> noise_mask. comfyanonymous 2024-11-19 15:31:09 -05:00
  • 156a28786b Add boolean to InpaintModelConditioning to disable the noise mask. comfyanonymous 2024-11-19 07:31:29 -05:00
  • f498d855ba Add terminal size fallback (#5623) Yoland Yan 2024-11-19 00:34:20 -08:00
  • b699a15062 Refactor inpaint/ip2p code. comfyanonymous 2024-11-19 02:34:35 -05:00
  • 9cc90ee3eb Update UI screenshot in README (#5666) Chenlei Hu 2024-11-18 16:50:34 -05:00
  • 9a0a5d32ee Add a skip layer guidance node that can also skip single layers. comfyanonymous 2024-11-18 02:20:43 -05:00
  • d9f90965c8 Support block replace patches in auraflow. comfyanonymous 2024-11-17 08:19:59 -05:00
  • 41886af138 Add transformer options blocks replace patch to mochi. comfyanonymous 2024-11-16 14:47:06 -05:00
  • 22a1d7ce78 Fix 3.8 compatibility in user_manager.py (#5645) Chenlei Hu 2024-11-16 20:42:21 -05:00
  • 4ac401af2b Update web content to release v1.3.44 (#5620) Chenlei Hu 2024-11-15 20:17:15 -05:00
  • 5fb59c8475 Add a node to block merge auraflow models. comfyanonymous 2024-11-15 12:47:55 -05:00
  • 122c9ca1ce Add advanced model merging node for mochi. comfyanonymous 2024-11-14 07:51:20 -05:00
  • 3b9a6cf2b1 Fix issue with 3d masks. comfyanonymous 2024-11-13 07:18:30 -05:00
  • 3748e7ef7a Fix regression. comfyanonymous 2024-11-13 04:24:48 -05:00
  • 8ebf2d8831 Add block replace transformer_options to flux. comfyanonymous 2024-11-12 08:00:00 -05:00
  • a72d152b0c fix --cuda-device arg for AMD/HIP devices (#5586) Bratzmeister 2024-11-12 11:53:36 +00:00
  • eb476e6ea9 Allow 1D masks for 1D latents. comfyanonymous 2024-11-11 14:44:19 -05:00
  • 2d28b0b479 improve: add descriptions for clip loaders (#5576) Dr.Lt.Data 2024-11-11 19:37:23 +09:00
  • 8b275ce5be Support auto detecting some zsnr anime checkpoints. comfyanonymous 2024-11-11 05:25:16 -05:00
  • 2a18e98ccf Refactor so that zsnr can be set in the sampling_settings. comfyanonymous 2024-11-11 04:55:56 -05:00
  • 8a5281006f Fix some custom nodes. comfyanonymous 2024-11-10 22:41:00 -05:00
  • bdeb1c171c Fast previews for mochi. comfyanonymous 2024-11-10 03:39:35 -05:00
  • 9c1ed58ef2 proper fix for sag. comfyanonymous 2024-11-10 00:10:45 -05:00
  • 8b90e50979 Properly handle and reshape masks when used on 3d latents. comfyanonymous 2024-11-09 07:10:43 -05:00
  • 6ee066a14f Live terminal output (#5396) pythongosssss 2024-11-09 00:13:34 +00:00
  • dd5b57e3d7 fix for SAG with Kohya HRFix/ Deep Shrink (#5546) DenOfEquity 2024-11-08 23:16:29 +00:00
  • 75a818c720 Move mochi latent node to: latent/video. comfyanonymous 2024-11-08 08:33:44 -05:00
  • 2865f913f7 Free memory before doing tiled decode. comfyanonymous 2024-11-07 04:01:24 -05:00
  • b49616f951 Make VAEDecodeTiled node work with video VAEs. comfyanonymous 2024-11-07 03:47:12 -05:00
  • 5e29e7a488 Remove scaled_fp8 key after reading it to silence warning. comfyanonymous 2024-11-06 04:56:42 -05:00
  • 8afb97cd3f Fix unknown VAE being detected as the mochi VAE. comfyanonymous 2024-11-05 03:42:58 -05:00
  • 69694f40b3 fix dynamic shape export (#5490) contentis 2024-11-04 11:59:28 -08:00
  • c49025f01b Allow POST /userdata/{file} endpoint to return full file info (#5446) Chenlei Hu 2024-11-04 13:57:21 -05:00
  • 696672905f Add mochi support to readme. v0.2.7 comfyanonymous 2024-11-04 04:55:07 -05:00
  • 6c9dbde7de Fix mochi all in one checkpoint t5xxl key names. comfyanonymous 2024-11-03 01:40:42 -05:00
  • ee8abf0cff Update folder paths: "clip" -> "text_encoders" comfyanonymous 2024-11-02 15:35:38 -04:00
  • fabf449feb Mochi VAE encoder. comfyanonymous 2024-11-01 17:33:09 -04:00
  • cc9cf6d1bd Rename some nodes in Display Name Mappings (nodes.py) (#5439) Uriel Deveaud 2024-10-31 13:18:05 -06:00
  • 1c8286a44b Avoid SyntaxWarning in UniPC docstring (#5442) Aarni Koskela 2024-10-31 21:17:26 +02:00
  • 1af4a47fd1 Bump up mac version for attention upcast bug workaround. comfyanonymous 2024-10-31 15:14:44 -04:00
  • f2aaa0a475 Rename ImageCrop to Image Crop (#5424) Uriel Deveaud 2024-10-30 22:35:34 -06:00
  • daa1565b93 Fix diffusers flux controlnet regression. comfyanonymous 2024-10-30 13:11:34 -04:00
  • 09fdb2b269 Support SD3.5 medium diffusers format weights and loras. comfyanonymous 2024-10-30 04:24:00 -04:00
  • 65a8659182 Update web content to release v1.3.26 (#5413) v0.2.6 Chenlei Hu 2024-10-29 14:14:06 -04:00
  • 770ab200f2 Cleanup SkipLayerGuidanceSD3 node. comfyanonymous 2024-10-29 10:11:46 -04:00
  • 954683d0db SLG first implementation for SD3.5 (#5404) Dango233 2024-10-29 21:59:21 +08:00
  • 30c0c81351 Add a way to patch blocks in SD3. v0.2.5 comfyanonymous 2024-10-29 00:48:32 -04:00
  • 13b0ff8a6f Update SD3 code. comfyanonymous 2024-10-28 21:58:52 -04:00
  • c320801187 Remove useless line. comfyanonymous 2024-10-28 17:41:12 -04:00
  • c0b0cfaeec Update web content to release v1.3.21 (#5351) Chenlei Hu 2024-10-28 14:29:38 -04:00
  • 669d9e4c67 Set default shift on mochi to 6.0 comfyanonymous 2024-10-27 22:21:04 -04:00
  • 9ee0a6553a float16 inference is a bit broken on mochi. comfyanonymous 2024-10-27 04:56:40 -04:00
  • 5cbb01bc2f Basic Genmo Mochi video model support. comfyanonymous 2024-10-26 06:54:00 -04:00
  • c3ffbae067 Make LatentUpscale nodes work on 3d latents. comfyanonymous 2024-10-26 01:50:51 -04:00
  • d605677b33 Make euler_ancestral work on flow models (credit: Ashen). comfyanonymous 2024-10-25 19:53:44 -04:00
  • ce759b7db6 Revert download to .tmp in frontend_management (#5369) Chenlei Hu 2024-10-25 19:26:13 -04:00
  • 52810907e2 Add a model merge node for SD3.5 large. comfyanonymous 2024-10-24 16:45:46 -04:00
  • af8cf79a2d support SimpleTuner lycoris lora for SD3 (#5340) PsychoLogicAu 2024-10-24 16:18:32 +11:00
  • 66b0961a46 Fix ControlLora issue with last commit. comfyanonymous 2024-10-23 17:02:19 -04:00
  • 754597c8a9 Clean up some controlnet code. comfyanonymous 2024-10-23 14:19:05 -04:00
  • 915fdb5745 Fix lowvram edge case. comfyanonymous 2024-10-22 16:34:50 -04:00
  • 5a8a48931a remove attention abstraction (#5324) contentis 2024-10-22 11:02:38 -07:00
  • 8ce2a1052c Optimizations to --fast and scaled fp8. v0.2.4 comfyanonymous 2024-10-22 02:12:28 -04:00
  • f82314fcfc Fix duplicate sigmas on beta scheduler. comfyanonymous 2024-10-21 20:19:45 -04:00
  • 0075c6d096 Mixed precision diffusion models with scaled fp8. comfyanonymous 2024-10-21 18:12:51 -04:00
  • 83ca891118 Support scaled fp8 t5xxl model. comfyanonymous 2024-10-20 22:27:00 -04:00
  • f9f9faface Fixed model merging issue with scaled fp8. comfyanonymous 2024-10-20 06:24:31 -04:00
  • 471cd3eace fp8 casting is fast on GPUs that support fp8 compute. comfyanonymous 2024-10-20 00:54:47 -04:00
  • a68bbafddb Support diffusion models with scaled fp8 weights. comfyanonymous 2024-10-19 23:47:42 -04:00
  • 73e3a9e676 Clamp output when rounding weight to prevent Nan. comfyanonymous 2024-10-19 19:07:10 -04:00
  • 518c0dc2fe Add tooltips to LoraSave node. comfyanonymous 2024-10-18 06:01:09 -04:00
  • ce0542e10b Add a note that python 3.13 is not yet supported to the README. comfyanonymous 2024-10-17 19:27:37 -04:00
  • 8473019d40 Pytorch can be shipped with numpy 2 now. comfyanonymous 2024-10-17 19:15:17 -04:00
  • 89f15894dd Ignore more network related errors during websocket communication. (#5269) Xiaodong Xie 2024-10-18 00:31:45 +02:00
  • 67158994a4 Use the lowvram cast_to function for everything. comfyanonymous 2024-10-17 17:25:56 -04:00
  • 7390ff3b1e Add missing import. comfyanonymous 2024-10-16 14:58:30 -04:00
  • 0bedfb26af Revert "Fix Transformers FutureWarning (#5140)" comfyanonymous 2024-10-16 12:36:19 -04:00
  • f71cfd2687 Add an experimental node to sharpen latents. comfyanonymous 2024-10-16 05:25:31 -04:00
  • c695c4af7f Frontend Manager: avoid redundant gh calls for static versions (#5152) Alex "mcmonkey" Goodwin 2024-10-16 00:35:37 -07:00
  • 0dbba9f751 Add some latent operation nodes. comfyanonymous 2024-10-15 15:00:36 -04:00
  • f584758271 Cleanup some useless lines. comfyanonymous 2024-10-14 21:02:29 -04:00
  • 95b7cf9bbe Fix Transformers FutureWarning (#5140) svdc 2024-10-14 21:12:20 -03:00
  • 191a0d56b4 Switch default packaging workflows to python 3.12 comfyanonymous 2024-10-13 06:59:31 -04:00
  • 3c60ecd7a8 Fix fp8 ops staying enabled. comfyanonymous 2024-10-12 14:10:13 -04:00