Commit Graph

  • 746a0410d4 Fix VAEEncode with taesd3. comfyanonymous 2024-06-16 03:10:04 -04:00
  • 04e8798c37 Improvements to the TAESD3 implementation. comfyanonymous 2024-06-16 02:04:24 -04:00
  • df7db0e027 support TAESD3 (#3738) Dr.Lt.Data 2024-06-16 15:03:53 +09:00
  • bb1969cab7 Initial support for the stable audio open model. comfyanonymous 2024-06-15 12:14:56 -04:00
  • 1281f933c1 Small optimization. comfyanonymous 2024-06-15 02:44:38 -04:00
  • f2e844e054 Optimize some unneeded if conditions in the sampling code. comfyanonymous 2024-06-15 02:25:03 -04:00
  • 0ec513d877 Add a --force-channels-last to inference models in channel last mode. comfyanonymous 2024-06-15 01:08:12 -04:00
  • 0e06b370db Print key names for easier debugging. comfyanonymous 2024-06-14 18:18:53 -04:00
  • 5eb98f0092 Exempt IPEX from non_blocking previews fixing segmentation faults. (#3708) Simon Lui 2024-06-13 15:51:14 -07:00
  • ac151ac169 Support SD3 diffusers lora. comfyanonymous 2024-06-13 18:26:01 -04:00
  • 37a08a41b3 Support setting weight offsets in weight patcher. comfyanonymous 2024-06-13 17:12:50 -04:00
  • 605e64f6d3 Fix lowvram issue. comfyanonymous 2024-06-12 10:39:33 -04:00
  • 0eaa34ec5b Fix regular empty latent image not working with SD3 and custom sampler. comfyanonymous 2024-06-12 10:32:34 -04:00
  • 321e509e0a Add link to SD3 example page to README. comfyanonymous 2024-06-12 09:48:27 -04:00
  • c8b5e08dc3 Default shift value on SD3 is 3.0 comfyanonymous 2024-06-12 02:24:39 -04:00
  • 1ddf512fdc Don't auto convert clip and vae weights to fp16 when saving checkpoint. comfyanonymous 2024-06-12 01:07:58 -04:00
  • 32be358213 Save SD3 modelspec.architecture in CheckpointSave node. comfyanonymous 2024-06-12 01:02:07 -04:00
  • 694e0b48e0 SD3 better memory usage estimation. comfyanonymous 2024-06-12 00:49:00 -04:00
  • 69c8d6d8a6 Single and dual clip loader nodes support SD3. comfyanonymous 2024-06-11 23:27:39 -04:00
  • 0e49211a11 Load the SD3 T5xxl model in the same dtype stored in the checkpoint. comfyanonymous 2024-06-11 17:03:26 -04:00
  • 5889b7ca0a Support multiple text encoder configurations on SD3. comfyanonymous 2024-06-11 13:14:43 -04:00
  • 1c34d338d7 Update EmptySD3LatentImage to use 1024 resolution by default. comfyanonymous 2024-06-11 07:37:22 -04:00
  • 9424522ead Reuse code. comfyanonymous 2024-06-11 07:20:26 -04:00
  • 73ce178021 Remove redundancy in mmdit.py (#3685) Dango233 2024-06-11 18:30:25 +08:00
  • 4134564dc1 Require safetensors library to be at least 0.4.2 for fp8 support. comfyanonymous 2024-06-11 06:26:13 -04:00
  • a82fae2375 Fix bug with cosxl edit model. comfyanonymous 2024-06-10 16:00:03 -04:00
  • 8c4a9befa7 SD3 Support. comfyanonymous 2024-06-10 13:26:25 -04:00
  • a5e6a632f9 Support sampling non 2D latents. comfyanonymous 2024-06-10 01:05:53 -04:00
  • 742d5720d1 Support zeroing out text embeddings with the attention mask. comfyanonymous 2024-06-09 16:41:04 -04:00
  • 6cd8ffc465 Reshape the empty latent image to the right amount of channels if needed. comfyanonymous 2024-06-08 02:16:55 -04:00
  • 56333d4850 Use the end token for the text encoder attention mask. comfyanonymous 2024-06-07 03:05:23 -04:00
  • 0dccb4617d Remove some unnecessary arguments. comfyanonymous 2024-06-06 14:49:45 -04:00
  • 104fcea0c8 Add function to get the list of currently loaded models. comfyanonymous 2024-06-05 19:14:56 -04:00
  • b1fd26fe9e pytorch xpu should be flash or mem efficient attention? comfyanonymous 2024-06-04 17:44:14 -04:00
  • 20447e9ec9 Fix alpha in PorterDuffImageComposite. (#3411) Denys Smirnov 2024-06-04 23:37:11 +03:00
  • cb8d0ebccc Don't load the view coordinates when loading a workflow from the history. comfyanonymous 2024-06-03 19:48:27 -04:00
  • 809cc85a8e Remove useless code. comfyanonymous 2024-06-02 19:21:53 -04:00
  • b249862080 Add an annoying print to a function I want to remove. comfyanonymous 2024-06-01 12:47:31 -04:00
  • e2c585f3be Fix to allow use of PerpNegGuider with cfg_function_post hooks (like PAG) (#3618) Peter Crabtree 2024-06-01 12:36:08 -04:00
  • 04b308229e Small refactor of preview code. comfyanonymous 2024-05-31 11:18:37 -04:00
  • bf3e334d46 Disable non_blocking when --deterministic or directml. comfyanonymous 2024-05-30 11:07:38 -04:00
  • 71ec5b144e Update commands to install nightly pytorch in readme. comfyanonymous 2024-05-29 00:19:30 -04:00
  • 91542d4f8b Import spandrel_extra_arches if present. comfyanonymous 2024-05-28 01:37:40 -04:00
  • b26da2245f Fix UnetParams annotation typo (#3589) JettHu 2024-05-28 07:30:35 +08:00
  • 0920e0e5fe Remove some unused imports. comfyanonymous 2024-05-27 19:03:56 -04:00
  • 34030fed92 improve dom widget performance (#3584) luke zhang 2024-05-28 02:26:07 +08:00
  • f6a203951f Extend core snapToGrid to LiteGraph Groups. (#3393) Regis Gaughan, III 2024-05-27 14:05:51 -04:00
  • 16a493a190 Keep compatibility with some custom nodes. comfyanonymous 2024-05-26 15:37:24 -04:00
  • 9a151b7def Fix issue and unpin spandrel package. comfyanonymous 2024-05-26 13:44:47 -04:00
  • 8cfd677cc0 Replace chainner_models with Spandrel package (#2146) Joey Ballentine 2024-05-26 13:44:17 -04:00
  • ffc4b7c30e Fix DORA strength. comfyanonymous 2024-05-25 02:31:23 -04:00
  • 5b87369474 Load titles from API format JSON (#3563) DLohn 2024-05-24 20:53:15 -07:00
  • efa5a711b2 Reduce memory usage when applying DORA: #3557 comfyanonymous 2024-05-24 23:36:48 -04:00
  • 58c9838274 Speed up TAESD preview. comfyanonymous 2024-05-24 02:37:57 -04:00
  • b02bcced05 Fix FreeU not working when shape is tensor. comfyanonymous 2024-05-23 11:47:43 -04:00
  • 6507a9c716 Remove the CTRL-Delete keybind. comfyanonymous 2024-05-23 01:29:22 -04:00
  • 6c23854f54 Fix OSX latent2rgb previews. comfyanonymous 2024-05-22 13:56:28 -04:00
  • 7718ada4ed Add type annotation UnetWrapperFunction (#3531) Chenlei Hu 2024-05-22 02:07:27 -04:00
  • 8508df2569 Work around black image bug on Mac 14.5 by forcing attention upcasting. comfyanonymous 2024-05-21 16:56:33 -04:00
  • 83d969e397 Disable xformers when tracing model. comfyanonymous 2024-05-21 13:55:49 -04:00
  • 1900e5119f Fix potential issue. comfyanonymous 2024-05-20 08:19:54 -04:00
  • 276f8fce9f Print error when node is missing. comfyanonymous 2024-05-20 07:03:06 -04:00
  • 4bc1884478 Provide a better error message when attempting to execute the workflow with a missing node. (#3517) Dr.Lt.Data 2024-05-20 19:58:46 +09:00
  • 09e069ae6c Log the pytorch version. comfyanonymous 2024-05-20 06:22:29 -04:00
  • 11a2ad5110 Fix controlnet not upcasting on models that have it enabled. comfyanonymous 2024-05-19 17:58:03 -04:00
  • 4ae1515f14 Slightly faster latent2rgb previews. comfyanonymous 2024-05-19 17:42:35 -04:00
  • f37a47110b Make --preview-method auto default to the fast latent2rgb previews. comfyanonymous 2024-05-19 11:45:36 -04:00
  • 0bdc2b15c7 Cleanup. comfyanonymous 2024-05-18 10:11:44 -04:00
  • 98f828fad9 Remove unnecessary code. comfyanonymous 2024-05-18 09:36:26 -04:00
  • 1c4af5918a Better error message if the webcam node doesn't work. comfyanonymous 2024-05-17 14:02:09 -04:00
  • 91590adf04 Add webcam node (#3497) pythongosssss 2024-05-17 18:16:08 +01:00
  • 19300655dd Don't automatically switch to lowvram mode on GPUs with low memory. comfyanonymous 2024-05-17 00:31:32 -04:00
  • 46daf0a9a7 Add debug options to force on and off attention upcasting. comfyanonymous 2024-05-16 04:09:41 -04:00
  • 58f8388020 More proper fix for #3484. comfyanonymous 2024-05-16 00:11:01 -04:00
  • 2d41642716 Fix lowvram dora issue. comfyanonymous 2024-05-15 02:40:06 -04:00
  • ec6f16adb6 Fix SAG. comfyanonymous 2024-05-14 18:02:27 -04:00
  • bb4940d837 Only enable attention upcasting on models that actually need it. comfyanonymous 2024-05-14 15:18:00 -04:00
  • b0ab31d06c Refactor attention upcasting code part 1. comfyanonymous 2024-05-14 12:47:31 -04:00
  • 2de3b69b30 Support saving some more modelspec types. comfyanonymous 2024-05-13 21:54:11 -04:00
  • cf6e1efb69 Show message on error when loading wf from file (works on drag and drop) (#3466) freakabcd 2024-05-14 05:22:22 +10:00
  • ece5acb8e8 Fix nightly package workflow. comfyanonymous 2024-05-12 16:05:10 -04:00
  • 794a357f7a Update the nightly workflow. comfyanonymous 2024-05-12 07:24:12 -04:00
  • 22edd3add5 Fix to LoadImage Node for #3416 HDR images loading additional smaller… (#3454) shawnington 2024-05-12 04:07:38 -07:00
  • f509c6fe21 Fix Intel GPU memory allocation accuracy and documentation update. (#3459) Simon Lui 2024-05-12 03:36:30 -07:00
  • fa6dd7e5bb Fix lowvram issue with saving checkpoints. comfyanonymous 2024-05-12 06:13:45 -04:00
  • 49c20cdc70 No longer necessary. comfyanonymous 2024-05-12 05:34:43 -04:00
  • e1489ad257 Fix issue with lowvram mode breaking model saving. comfyanonymous 2024-05-11 21:46:05 -04:00
  • 4f63ee99f1 Add a button to reset the view. comfyanonymous 2024-05-10 17:30:52 -04:00
  • f374ea714d Setting for saving and restoring canvas position and zoom level (#3437) pythongosssss 2024-05-10 22:07:46 +01:00
  • 0fecfd2b1a Added generic wrapper function node_helpers.pillow to fix PIL issues #4472 and #2445 (#3422) shawnington 2024-05-09 02:38:00 -07:00
  • 93e876a3be Remove warnings that confuse people. comfyanonymous 2024-05-09 04:39:46 -04:00
  • cd07340d96 Typo fix. comfyanonymous 2024-05-08 18:36:56 -04:00
  • c33412288f Fix issue with loading some JPG: #3416 comfyanonymous 2024-05-07 05:41:06 -04:00
  • d7fa417bfa feat: shortcuts for zoom in/out (#3410) Dr.Lt.Data 2024-05-07 17:40:56 +09:00
  • c61eadf69a Make the load checkpoint with config function call the regular one. comfyanonymous 2024-05-06 20:04:39 -04:00
  • 3787b4f246 Use get_model_object in Deep Shrink node (#3408) Pam 2024-05-07 03:39:39 +05:00
  • 565eb6d176 Add a SplitSigmasDenoise node as an alternative to SplitSigmas. comfyanonymous 2024-05-05 05:24:36 -04:00
  • 9a70b70de4 add opacity slider in maskeditor (#3404) vilanele 2024-05-05 11:01:06 +02:00
  • 72508a8d19 Only set LOAD_TRUNCATED_IMAGES when if the Image open fails. comfyanonymous 2024-05-04 03:39:37 -04:00
  • 0d45efb7d6 Fixed Issue with LoadImage node when loading PNG files with embedded ICC profiles. (#3316) shawnington 2024-05-04 02:32:41 -05:00