Commit Graph

Select branches

Hide Pull Requests

master

Mono Color

• e4f7ea105f Added context window support to core sampling code (#9238) Jedrzej Kosinski 2025-08-13 18:33:05 -07:00
• c991a5da65 Fix XPU iGPU regressions (#9322) Simon Lui 2025-08-13 16:13:35 -07:00
• 9df8792d4b Make last PR not crash comfy on old pytorch. (#9324) comfyanonymous 2025-08-13 12:12:41 -07:00
• 3da5a07510 SDPA backend priority (#9299) contentis 2025-08-13 20:53:27 +02:00
• afa0a45206 Reduce portable size again. (#9323) comfyanonymous 2025-08-13 11:42:08 -07:00
• 615eb52049 Put back frontend version. (#9317) comfyanonymous 2025-08-13 00:48:06 -07:00
• d5c1954d5c ComfyUI version 0.3.50 comfyanonymous 2025-08-13 03:46:38 -04:00
• e400f26c8f Downgrade frontend for release. (#9316) comfyanonymous 2025-08-13 00:44:54 -07:00
• 5ca8e2fac3 Update release workflow to python3.13 pytorch cu129 (#9315) comfyanonymous 2025-08-13 00:01:12 -07:00
• 3294782d19 Update template to 0.1.59 (#9313) ComfyUI Wiki 2025-08-13 14:50:50 +08:00
• 898d88e10e Make torchaudio exception catching less specific (#9309) Jedrzej Kosinski 2025-08-12 20:34:58 -07:00
• 560d38f34c Wan2.2 fun control support. (#9292) comfyanonymous 2025-08-12 20:26:33 -07:00
• e1d4f36d8d Update test release package workflow with python 3.13 cu129. (#9306) comfyanonymous 2025-08-12 17:13:04 -07:00
• 1e3ae1eed8 Update template to 0.1.58 (#9302) ComfyUI Wiki 2025-08-13 05:14:27 +08:00
• f4231a80b1 fix(Kling Image API Node): do not pass "image_type" when no image (#9271) Alexander Piskun 2025-08-12 00:15:14 +03:00
• 2208aa616d Support SimpleTuner lycoris lora for Qwen-Image (#9280) PsychoLogicAu 2025-08-12 06:56:16 +10:00
• 629b173837 Update template & embedded docs (#9283) ComfyUI Wiki 2025-08-12 04:52:12 +08:00
• fa340add55 remove creation of non-used asyncio_loop (#9284) Alexander Piskun 2025-08-11 23:48:17 +03:00
• 966f3a5206 Only show feature flags log when verbose. (#9281) comfyanonymous 2025-08-11 02:53:01 -07:00
• 0552de7c7d Bump pytorch cuda and rocm versions in readme instructions. (#9273) comfyanonymous 2025-08-10 02:03:47 -07:00
• 5828607ccf Not sure if AMD actually support fp16 acc but it doesn't crash. (#9258) comfyanonymous 2025-08-09 09:49:25 -07:00
• 735bb4bdb1 Users report gfx1201 is buggy on flux with pytorch attention. (#9244) comfyanonymous 2025-08-08 01:21:00 -07:00
• bf2a1b5b1e async API nodes (#9129) Alexander Piskun 2025-08-08 06:37:50 +03:00
• 42974a448c _ui.py import torchaudio safety check (#9234) Jedrzej Kosinski 2025-08-07 14:54:09 -07:00
• 05df2df489 Fix RepeatLatentBatch not working on multi dim latents. (#9227) comfyanonymous 2025-08-07 08:20:40 -07:00
• 37d620a6b8 Update frontend to v1.24.3 (#9175) Christian Byrne 2025-08-06 16:52:39 -07:00
• 32691b16f4 Update template to 0.1.52 (#9206) ComfyUI Wiki 2025-08-07 01:26:29 +08:00
• 4c3e57b0ae Fixed an issue where qwenLora could not be loaded properly. (#9208) flybirdxx 2025-08-07 01:23:11 +08:00
• 9126c0cfe4 Qwen Image model merging node. (#9202) comfyanonymous 2025-08-06 01:07:04 -07:00
• d8c51ba15a Add Qwen Image model to readme. (#9191) comfyanonymous 2025-08-05 04:41:18 -07:00
• 32a95bba8a ComfyUI version 0.3.49 comfyanonymous 2025-08-05 07:33:02 -04:00
• da1ad9b516 Update template to 0.1.51 (#9187) ComfyUI Wiki 2025-08-05 19:24:12 +08:00
• d044a24398 Fix default shift and any latent size for qwen image model. (#9186) comfyanonymous 2025-08-05 03:12:27 -07:00
• 5be6fd09ff Update template to 0.1.48 (#9182) ComfyUI Wiki 2025-08-05 15:48:56 +08:00
• f69609bbd6 Add Veo3 video generation node with audio support (#9110) Christian Byrne 2025-08-04 22:52:25 -07:00
• c012400240 Initial support for qwen image model. (#9179) comfyanonymous 2025-08-04 19:53:25 -07:00
• 03895dea7c Fix another issue with the PR. (#9170) comfyanonymous 2025-08-04 01:33:04 -07:00
• 84f9759424 Add some warnings and prevent crash when cond devices don't match. (#9169) comfyanonymous 2025-08-04 01:20:12 -07:00
• 7991341e89 Various fixes for broken things from earlier PR. (#9168) comfyanonymous 2025-08-04 01:02:40 -07:00
• 140ffc7fdc Fix broken controlnet from last PR. (#9167) comfyanonymous 2025-08-04 00:28:12 -07:00
• 182f90b5ec Lower cond vram use by casting at the same time as device transfer. (#9159) comfyanonymous 2025-08-04 00:11:53 -07:00
• aebac22193 Cleanup. (#9160) comfyanonymous 2025-08-03 04:08:11 -07:00
• 13aaa66ec2 Make sure context is on the right device. (#9154) comfyanonymous 2025-08-02 12:09:23 -07:00
• 5f582a9757 Make sure all the conds are on the right device. (#9151) comfyanonymous 2025-08-02 12:00:13 -07:00
• fbcc23945d Update template to 0.1.47 (#9153) ComfyUI Wiki 2025-08-03 02:15:29 +08:00
• 3dfefc88d0 API for Recently Used Items (#8792) Johnpaul Chiwetelu 2025-08-02 03:02:06 +01:00
• bff60b5cfc ComfyUI version 0.3.48 comfyanonymous 2025-08-01 20:03:22 -04:00
• 1e638a140b Tiny wan vae optimizations. (#9136) comfyanonymous 2025-08-01 02:25:38 -07:00
• 4696d74305 update template to 0.1.45 (#9135) ComfyUI Wiki 2025-08-01 15:06:18 +08:00
• 5ee381c058 Fix WanFirstLastFrameToVideo node when no clip vision. (#9134) comfyanonymous 2025-07-31 20:33:27 -07:00
• 4887743a2a V3 Node Schema Definition - initial (#8656) Jedrzej Kosinski 2025-07-31 15:02:12 -07:00
• 97b8a2c26a More accurate explanation of release process. (#9126) comfyanonymous 2025-07-31 02:46:23 -07:00
• 97eb256a35 Add support for partial execution in backend (#9123) guill 2025-07-30 19:55:28 -07:00
• 61b08d4ba6 Replace manual x * sigmoid(x) with torch silu in VAE nonlinearity (#9057) chaObserv 2025-07-31 07:25:56 +08:00
• da9dab7edd Small wan camera memory optimization. (#9111) comfyanonymous 2025-07-30 02:55:26 -07:00
• d2aaef029c Update template to 0.1.44 (#9104) ComfyUI Wiki 2025-07-30 10:50:49 +08:00
• 0a3d062e06 ComfyAPI Core v0.0.2 (#8962) guill 2025-07-29 19:17:22 -07:00
• 2f74e17975 ComfyUI version 0.3.47 comfyanonymous 2025-07-29 20:08:25 -04:00
• dca6bdd4fa Make wan2.2 5B i2v take a lot less memory. (#9102) comfyanonymous 2025-07-29 16:44:18 -07:00
• 7d593baf91 Extra reserved vram on large cards on windows. (#9093) comfyanonymous 2025-07-29 01:07:45 -07:00
• c60dc4177c Remove unecessary clones in the wan2.2 VAE. (#9083) comfyanonymous 2025-07-28 11:48:19 -07:00
• 5d4cc3ba1b ComfyUI 0.3.46 comfyanonymous 2025-07-28 08:04:04 -04:00
• 9f1388c0a3 Add wan2.2 to readme. (#9081) comfyanonymous 2025-07-28 05:01:53 -07:00
• a88788dce6 Wan 2.2 support. (#9080) comfyanonymous 2025-07-28 05:00:23 -07:00
• d0210fe2e5 Update template to 0.1.41 (#9079) ComfyUI Wiki 2025-07-28 19:55:02 +08:00
• e6d9f62744 Add Moonvalley Marey V2V node with updated input validation (#9069) Christian Byrne 2025-07-27 16:51:36 -07:00
• 78672d0ee6 Small readme update. (#9071) comfyanonymous 2025-07-27 04:42:58 -07:00
• 1ef70fcde4 Fix the broken link (#9060) ComfyUI Wiki 2025-07-27 05:25:33 +08:00
• 0621d73a9c Remove useless code. (#9059) comfyanonymous 2025-07-26 01:44:19 -07:00
• b850d9a8bb Add map_function to get_history. (#9056) comfyanonymous 2025-07-25 18:25:45 -07:00
• c60467a148 Update negative prompt for Moonvalley nodes (#9038) Thor-ATX 2025-07-26 09:27:03 +12:00
• c0207b473f Fix issue with line endings github workflow. (#9053) comfyanonymous 2025-07-25 14:25:08 -07:00
• 93bc2f8e4d Update template to 0.1.40 (#9048) ComfyUI Wiki 2025-07-26 01:24:23 +08:00
• e6e5d33b35 Remove useless code. (#9041) comfyanonymous 2025-07-25 01:58:28 -07:00
• 4293e4da21 Add WAN ATI support (#8874) Eugene Fairley 2025-07-24 17:59:19 -07:00
• 69cb57b342 Print xpu device name. (#9035) comfyanonymous 2025-07-24 12:06:25 -07:00
• d03ae077b4 Added parameter required_frontend_version in the /system_stats API response (#8875) SHIVANSH GUPTA 2025-07-24 23:35:54 +05:30
• 0ccc88b03f Support Iluvatar CoreX (#8585) honglyua 2025-07-25 01:57:36 +08:00
• eb2f78b4e0 [Training Node] algo support, grad acc, optional grad ckpt (#9015) Kohaku-Blueleaf 2025-07-24 08:57:27 +08:00
• e729a5cc11 Separate denoised and noise estimation in Euler CFG++ (#9008) chaObserv 2025-07-24 07:47:05 +08:00
• e78d230496 Only enable cuda malloc on cuda torch. (#9031) comfyanonymous 2025-07-23 16:37:43 -07:00
• d3504e1778 Enable pytorch attention by default for gfx1201 on torch 2.8 (#9029) comfyanonymous 2025-07-23 16:21:29 -07:00
• a86a58c308 Fix xpu function not implemented p2. (#9027) comfyanonymous 2025-07-23 15:18:20 -07:00
• 39dda1d40d Fix xpu function not implemented. (#9026) comfyanonymous 2025-07-23 15:10:59 -07:00
• 5ad33787de Add default device argument. (#9023) comfyanonymous 2025-07-23 11:20:49 -07:00
• 255f139863 Add xpu version for async offload and some other things. (#9004) Simon Lui 2025-07-22 12:20:09 -07:00
• 5ac9ec214b Try to fix line endings workflow. (#9001) comfyanonymous 2025-07-22 01:07:51 -07:00
• 0aa1c58b04 This is not needed. (#8991) comfyanonymous 2025-07-21 13:48:25 -07:00
• 5249e45a1c Add hidream e1.1 example to readme. (#8990) comfyanonymous 2025-07-21 12:23:41 -07:00
• 54a45b9967 Replace torchaudio.load with pyav. (#8989) comfyanonymous 2025-07-21 11:19:14 -07:00
• 9a470e073e ComfyUI version 0.3.45 comfyanonymous 2025-07-21 14:05:43 -04:00
• 7d627f764c Update template to 0.1.39 (#8981) ComfyUI Wiki 2025-07-21 03:58:35 +08:00
• a0c0785635 Document what the fast_fp16_accumulation is in the portable. (#8973) comfyanonymous 2025-07-19 22:24:09 -07:00
• 100c2478ea Add SamplingPercentToSigma node (#8963) chaObserv 2025-07-20 11:09:11 +08:00
• 1da5639e86 Update template to 0.1.37 (#8967) ComfyUI Wiki 2025-07-19 18:08:00 +08:00
• 1b96fae1d4 Add nested style of dual cfg to DualCFGGuider node. (#8965) comfyanonymous 2025-07-19 01:55:23 -07:00
• 7f492522b6 Forgot this (#8957) comfyanonymous 2025-07-18 02:43:02 -07:00
• 650838fd6f Experimental CFGNorm node. (#8942) comfyanonymous 2025-07-17 01:11:07 -07:00
• 491fafbd64 Silence clip tokenizer warning. (#8934) comfyanonymous 2025-07-16 11:42:07 -07:00
• 9bc2798f72 LTXV VAE decoder: switch default padding mode (#8930) Harel Cain 2025-07-16 19:54:38 +02:00