Dima/ComfyUI
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Commit Graph

  • e03fe8b591 Update command to install AMD stable linux pytorch. (#12437) comfyanonymous 2026-02-12 20:29:12 -08:00
  • ae79e33345 llama: use a more efficient rope implementation (#12434) rattus 2026-02-12 16:56:42 -08:00
  • 117e214354 ModelPatcherDynamic: force load non leaf weights (#12433) rattus 2026-02-12 16:51:50 -08:00
  • 4a93a62371 fix(api-nodes): add separate retry budget for 429 rate limit responses (#12421) Alexander Piskun 2026-02-12 11:38:51 +02:00
  • 66c18522fb Add a tip for common error. (#12414) comfyanonymous 2026-02-11 19:12:16 -08:00
  • e5ae670a40 Update ace15.py to allow min_p sampling (#12373) askmyteapot 2026-02-12 11:28:48 +10:00
  • 3fe61cedda model_patcher: guard against none model_dtype (#12410) rattus 2026-02-11 11:54:02 -08:00
  • 2a4328d639 ace15: Use dynamic_vram friendly trange (#12409) rattus 2026-02-11 11:53:42 -08:00
  • d297a749a2 dynamic_vram: Fix windows Aimdo crash + Fix LLM performance (#12408) rattus 2026-02-11 11:50:16 -08:00
  • 2b7cc7e3b6 [API Nodes] enable Magnific Upscalers (#12179) Alexander Piskun 2026-02-11 21:30:19 +02:00
  • 4993411fd9 Dispatch desktop auto-bump when a ComfyUI release is published (#12398) Benjamin Lu 2026-02-11 11:15:13 -08:00
  • 2c7cef4a23 fix(api-nodes): retry on connection errors during polling instead of aborting (#12393) Alexander Piskun 2026-02-11 20:51:49 +02:00
  • 76a7fa96db Make built in lora training work on anima. (#12402) comfyanonymous 2026-02-10 19:04:32 -08:00
  • cdcf4119b3 [Trainer] training with proper offloading (#12189) Kohaku-Blueleaf 2026-02-11 10:45:19 +08:00
  • dbe70b6821 Add a VideoSlice node (#12107) AustinMroz 2026-02-10 14:42:21 -08:00
  • 00fff6019e feat(jobs): add 3d to PREVIEWABLE_MEDIA_TYPES for first-class 3D output support (#12381) guill 2026-02-10 14:37:14 -08:00
  • 123a7874a9 ops: Fix vanilla-fp8 loaded lora quality (#12390) rattus 2026-02-10 10:38:28 -08:00
  • f719f9c062 sd: delay VAE dtype archive until after override (#12388) rattus 2026-02-10 10:37:46 -08:00
  • fe053ba5eb mp: dont deep-clone objects from model_options (#12382) rattus 2026-02-10 10:37:17 -08:00
  • 6648ab68bc ComfyUI v0.13.0 v0.13.0 comfyanonymous 2026-02-10 13:26:29 -05:00
  • 6615db925c chore: update workflow templates to v0.8.38 (#12394) ComfyUI Wiki 2026-02-11 02:24:56 +08:00
  • 8ca842a8ed feat(api-nodes-Kling): add new models (V3, O3) (#12389) Alexander Piskun 2026-02-10 19:34:54 +02:00
  • c1b63a7e78 fix(Moonvalley-API-Nodes): adjust "steps" parameter to not raise exception (#12370) Alexander Piskun 2026-02-10 04:58:27 +02:00
  • 349a636a2b chore: update workflow templates to v0.8.37 (#12377) ComfyUI Wiki 2026-02-10 10:25:34 +08:00
  • a4be04c5d7 Ace step prompts match now. (#12376) comfyanonymous 2026-02-09 16:45:56 -08:00
  • baf8c87455 Iimprovements to ACE-Steps 1.5 text encoding (part 2) (#12350) blepping 2026-02-09 17:41:49 -07:00
  • 62315fbb15 Dynamic VRAM fixes - Ace 1.5 performance + a VRAM leak (#12368) rattus 2026-02-09 13:16:08 -08:00
  • a0302cc6a8 Make tonemap latent work on any dim latents. (#12363) comfyanonymous 2026-02-08 18:16:40 -08:00
  • f350a84261 Disable prompt weights for ltxv2. (#12354) comfyanonymous 2026-02-07 16:16:28 -08:00
  • 3760d74005 chore: update embedded docs to v0.4.1 (#12346) ComfyUI Wiki 2026-02-08 07:34:52 +08:00
  • 9bf5aa54db Add search_aliases to sa-solver and seeds-2 node (#12327) chaObserv 2026-02-08 06:38:51 +08:00
  • 5ff4fdedba Fix LazyCache (#12344) Jukka Seppänen 2026-02-07 21:25:30 +02:00
  • 17e7df43d1 Pad ace step 1.5 ref audio if not long enough. (#12341) comfyanonymous 2026-02-06 21:02:11 -08:00
  • 039955c527 Some fixes to previous pr. (#12339) comfyanonymous 2026-02-06 17:14:52 -08:00
  • 6a26328842 Support fp16 for Cosmos-Predict2 and Anima (#12249) tdrussell 2026-02-06 19:12:15 -06:00
  • 204e65b8dc Fix bug with last pr (#12338) comfyanonymous 2026-02-06 16:48:20 -08:00
  • a831c19b70 Fix return_word_ids=True with Anima tokenizer (#12328) asagi4 2026-02-07 02:38:04 +02:00
  • eba6c940fd Make ace step 1.5 base model work properly with default workflow. (#12337) comfyanonymous 2026-02-06 16:14:56 -08:00
  • a1c101f861 EasyCache: Support LTX2 (#12231) Jukka Seppänen 2026-02-06 07:43:09 +02:00
  • c2d7f07dbf Fix issue when using disable_unet_model_creation (#12315) comfyanonymous 2026-02-05 16:24:09 -08:00
  • 458292fef0 Fix some lowvram stuff with ace step 1.5 (#12312) comfyanonymous 2026-02-05 16:15:04 -08:00
  • 6555dc65b8 Make ace step 1.5 work without the llm. (#12311) comfyanonymous 2026-02-05 13:43:45 -08:00
  • 2b70ab9ad0 Add a Create List node (#12173) AustinMroz 2026-02-04 22:18:21 -08:00
  • 00efcc6cd0 Bump comfyui-frontend-package to 1.38.13 (#12238) Comfy Org PR Bot 2026-02-05 15:17:37 +09:00
  • cb459573c8 ComfyUI v0.12.3 v0.12.3 comfyanonymous 2026-02-05 01:13:35 -05:00
  • 35183543e0 Add VAE tiled decode node for audio. (#12299) comfyanonymous 2026-02-04 22:12:04 -08:00
  • a246cc02b2 Improvements to ACE-Steps 1.5 text encoding (#12283) blepping 2026-02-04 22:17:37 -07:00
  • a50c32d63f Disable sage attention on ace step 1.5 (#12297) comfyanonymous 2026-02-04 19:15:30 -08:00
  • 6125b80979 Add llm sampling options and make reference audio work on ace step 1.5 (#12295) comfyanonymous 2026-02-04 18:29:22 -08:00
  • c8fcbd66ee Try to fix ace text encoder slowness on some configs. (#12290) comfyanonymous 2026-02-04 16:37:05 -08:00
  • 26dd7eb421 Fix ace step nan issue on some hardware/pytorch configs. (#12289) comfyanonymous 2026-02-04 15:25:06 -08:00
  • e77b34dfea add File3DAny output to Load3D node; extend SaveGLB to accept File3DAny as input (#12276) Alexander Piskun 2026-02-04 21:35:38 +02:00
  • ef73070ea4 mp: Fix checkpoint saving (#12268) rattus 2026-02-03 23:08:45 -08:00
  • d30c609f5a utils: safetensors: dont slice data on torch level (#12266) rattus 2026-02-03 22:48:47 -08:00
  • 5087f1d497 ComfyUI v0.12.2 v0.12.2 comfyanonymous 2026-02-04 00:08:59 -05:00
  • a31681564d Fix crash with ace step 1.5 (#12264) comfyanonymous 2026-02-03 21:03:21 -08:00
  • 855849c658 mm: Remove Aimdo exemption for empty_cache (#12260) rattus 2026-02-03 18:39:19 -08:00
  • fe2511468d Support the 4B ace step 1.5 lm model. (#12257) comfyanonymous 2026-02-03 16:01:38 -08:00
  • 3be0175166 ComfyUI v0.12.1 v0.12.1 comfyanonymous 2026-02-03 14:13:18 -05:00
  • b8315e66cb Fix tiled vae for ace step 1.5 (#12253) comfyanonymous 2026-02-03 11:40:45 -08:00
  • ab1050bec3 Support ace step 1.5 base model loras. (#12252) comfyanonymous 2026-02-03 10:54:23 -08:00
  • fb23935c11 feat(comfy_api): add basic 3D Model file types (#12129) Alexander Piskun 2026-02-03 20:31:46 +02:00
  • 85fc35e8fa Fix mac issue. (#12250) comfyanonymous 2026-02-03 09:19:39 -08:00
  • 223364743c llama: cast logits as a comfy-weight (#12248) comfyanonymous 2026-02-03 08:31:36 -08:00
  • affe881354 Fix some issues with mac. (#12247) comfyanonymous 2026-02-03 08:07:04 -08:00
  • f5030e26fd Add progress bar to ace step. (#12242) comfyanonymous 2026-02-03 01:09:30 -08:00
  • 66e1b07402 ComfyUI v0.12.0 v0.12.0 comfyanonymous 2026-02-03 02:20:59 -05:00
  • be4345d1c9 chore: update workflow templates to v0.8.31 (#12239) ComfyUI Wiki 2026-02-03 15:08:43 +08:00
  • 3c1a1a2df8 Basic support for the ace step 1.5 model. (#12237) comfyanonymous 2026-02-02 21:06:18 -08:00
  • ba5bf3f1a8 [API Nodes] HitPaw API nodes (#12117) Alexander Piskun 2026-02-03 05:17:59 +02:00
  • c05a08ae66 Add back function. (#12234) comfyanonymous 2026-02-02 16:52:07 -08:00
  • de9ada6a41 Dynamic VRAM unloading fix (#12227) rattus 2026-02-02 14:35:20 -08:00
  • 37f711d4a1 mm: Fix cast buffers with intel offloading (#12229) rattus 2026-02-02 14:34:46 -08:00
  • dd86b15521 Enable embeddings for some qwen 3 models. (#12218) comfyanonymous 2026-02-02 00:51:09 -08:00
  • 021ba20719 Fix issue with parameters on root model object. (#12216) comfyanonymous 2026-02-01 17:12:52 -08:00
  • b60be02aaf requirements: bump comfy-aimdo to 0.1.7 (#12211) rattus 2026-02-01 17:10:15 -08:00
  • 2b5da3b72e dynamic_vram: silence pytorch buffer warning (#12210) rattus 2026-02-01 17:09:55 -08:00
  • 794d05bdb1 dynamic_vram: respect argument cast dtypes in non-comfy weights (#12209) rattus 2026-02-01 17:09:21 -08:00
  • 361b9a82a3 fix pinning with model defined dtype (#12208) rattus 2026-02-01 08:42:32 -08:00
  • 667a1b8878 Fix some custom nodes breaking. (#12203) comfyanonymous 2026-01-31 22:55:18 -08:00
  • 32621c6a11 fix: improve error message when node type is missing (#12194) Christian Byrne 2026-01-31 22:13:48 -08:00
  • f8acd9c402 Reduce RAM usage, fix VRAM OOMs, and fix Windows shared memory spilling with adaptive model loading (#11845) rattus 2026-01-31 22:01:11 -08:00
  • 873de5f37a KV cache implementation for using llama models for text generation. (#12195) comfyanonymous 2026-01-31 18:11:11 -08:00
  • aa6f7a83bb Send is_input_list on v1 and v3 schema to frontend (#12188) Jedrzej Kosinski 2026-01-31 17:05:11 -08:00
  • 6ea8c128a3 Assets Part 2 - add more endpoints (#12125) Jedrzej Kosinski 2026-01-30 23:22:05 -08:00
  • 6e469a3f35 feat(api-nodes): add Q3 models and support for Extend and MultiFrame Vidu endpoints (#12175) Alexander Piskun 2026-01-31 08:44:08 +02:00
  • b8f848bfe3 Fix model not working with any res. (#12186) comfyanonymous 2026-01-30 21:12:48 -08:00
  • 4064062e7d Update python patch version in dep workflow. (#12184) comfyanonymous 2026-01-30 17:20:06 -08:00
  • 8aabe2403e Add color type and Color to RGB Int node (#12145) pythongosssss 2026-01-30 15:01:33 -08:00
  • 0167653781 feat(api-nodes): add RecraftCreateStyleNode node (#12055) Alexander Piskun 2026-01-31 00:04:43 +02:00
  • 0a7993729c Remove NodeInfoV3-related code; we are almost 100% guaranteed to stick with NodeInfoV1 for the foreseable future (#12147) Jedrzej Kosinski 2026-01-30 10:21:48 -08:00
  • bbe2c13a70 Make empty hunyuan latent 1.0 work with the 1.5 model. (#12171) comfyanonymous 2026-01-29 20:52:22 -08:00
  • 3aace5c8dc fix: count non-dict items in outputs_count (#12166) Christian Byrne 2026-01-29 17:10:08 -08:00
  • b0d9708974 ComfyUI v0.11.1 v0.11.1 comfyanonymous 2026-01-29 00:27:23 -05:00
  • c9b633d84f Add missing spacial downscale ratios. (#12146) comfyanonymous 2026-01-28 17:52:51 -08:00
  • 1711020904 chore: update workflow templates to v0.8.27 (#12141) ComfyUI Wiki 2026-01-29 01:48:02 +08:00
  • d9b8567547 bump manager version to 4.1b1 (#12140) Dr.Lt.Data 2026-01-29 02:47:37 +09:00
  • 6c5f906bf2 feat(api-nodes): add Grok Imagine nodes (#12136) Alexander Piskun 2026-01-28 19:46:57 +02:00
  • 4f5bd39b1c Update Python 3.14 compatibility notes in README (#12127) comfyanonymous 2026-01-27 16:58:48 -08:00
  • dcff27fe3f Add support for dev-only nodes. (#12106) guill 2026-01-27 13:03:29 -08:00