5138 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
Tara Ding	b9489e665d	Merge 875bdc4015 into 38ecad8f8a	2026-04-30 16:30:14 +08:00
Alexander Piskun	38ecad8f8a	feat(api-nodes): allow custom resolutions for GPTImage2 node (#13631) ... Signed-off-by: bigcat88 <bigcat88@icloud.com>	2026-04-30 01:09:33 -07:00
Jedrzej Kosinski	a7d82baa06	Fix SQLAlchemy version format in requirements.txt (#13547) ... Change SQLAlchemy>=2.0 to SQLAlchemy>=2.0.0 to satisfy the X.Y.Z version format expected by install_util.is_valid_version().	2026-04-29 23:30:01 -04:00
comfyanonymous	d10fc2d652	Lower peak mem usage for 8 bit formats with pyav. (#13626)	2026-04-29 23:05:31 -04:00
blepping	a164c82913	Add high quality preview support for Flux2 latents (#13496)	2026-04-29 19:37:30 -04:00
Talmaj	5eeae3f1d8	Cogvideox (#13402) ... --------- Co-authored-by: kijai <40791699+kijai@users.noreply.github.com> Co-authored-by: Talmaj Marinc <talmaj@comfy.org>	2026-04-29 19:30:08 -04:00
Jukka Seppänen	0e25a6936e	Reduce video tiny VAE peak VRAM and decode time (CORE-127) (#13617) ... * Update taehv.py * Simplify * Simplify pixel_unshuffle dispatch	2026-04-29 12:15:10 -07:00
rattus	fce0398470	dynamicVRAM + --cache-ram 2 (CORE-117) (#13603) ... * pinned_memory: remove JIT RAM pressure release This doesn't work, as freeing intermediates for pins needs to be higher-priority than freeing pins-for-pins if and when you are going to do that. So this is too late as pins-for-pins is model load time and we dont have JIT pins-for-pins. * cacheing: Add a filter to only free intermediates from inactive wfs This is to get priorities in amongst pins straight. * mm: free inactive-ram from RAM cache first Stuff from inactive workflows should be freed before anything else. * caching: purge old ModelPatchers first Dont try and score them, just dump them at the first sign of trouble if they arent part of the workflow.	2026-04-28 19:15:02 -04:00
comfyanonymous	dae3d34751	Use pyav to load images instead of pillow. (#13594) ... On failure (ex: animated webp files) fallback to old pillow code. This should fix the extra precision in high bit depth images (like 16 bit PNG) being discarded when loaded by Pillow and potentially add support for more image formats.	2026-04-28 18:15:06 -04:00
comfyanonymous	c7a517c2f9	Make pyav loading code handle tRNS PNG. (#13607)	2026-04-28 17:59:55 -04:00
rattus	e514119e1e	comfy-aimdo v0.3.0 (#13604) ... Comfy-aimdo 0.3.0 contains several major new features. multi-GPU support ARM support AMD support Refactorings include: Linkless architecture - linkage is now performed purely at runtime to stop host library lookups completely and only interact with the torch-loaded Nvidia stack. Elimination of cudart integration on linux. Its no consistent with windows. Misc bugfixes and minor features.	2026-04-28 16:34:37 -04:00
comfyanonymous	13519934ba	Handle metadata rotation in pyav code. (#13605)	2026-04-28 16:27:42 -04:00
Gilad Schreiber	24de8dc01b	Fix SolidMask and MaskComposite device mismatch with --gpu-only (#13296) ... SolidMask had a hardcoded device="cpu" while other nodes (e.g. EmptyImage) follow intermediate_device(). This causes a RuntimeError when MaskComposite combines masks from different device sources under --gpu-only. - SolidMask: use intermediate_device() instead of hardcoded "cpu" - MaskComposite: align source device to destination before operating Co-authored-by: Alexis Rolland <alexisrolland@hotmail.com> Co-authored-by: Jedrzej Kosinski <kosinkadink1@gmail.com>	2026-04-28 01:18:19 -07:00
Daxiong (Lin)	c0d77a5d53	Change the save 3d model node's filename prefix to 3d/ComfyUI (CORE-106) (#12826) ... * Change save 3d model's filename prefix to 3d/ComfyUI As this node has already changed from `Save GLB` to `Save 3D Model`, using the filename prefix `3d` will be better than `mesh` * use lowercase ---------	2026-04-28 00:59:59 -07:00
Tara Ding	875bdc4015	Update README	2026-04-27 23:05:42 -07:00
Tara Ding	1d64200d2e	Add benchmark README	2026-04-27 23:01:37 -07:00
Tara Ding	9ea25780c6	fix benchmark	2026-04-27 22:44:00 -07:00
Tara Ding	a2204ec976	force to regenerate prompts everytime	2026-04-27 22:34:38 -07:00
Tara Ding	54ced2923b	don't generate synthetic when load vbench fails	2026-04-27 22:28:42 -07:00
Tara Ding	c39f7ea76c	add tqdm to the benchmark	2026-04-27 22:23:42 -07:00
Tara Ding	ba978bc0e2	Fix format	2026-04-27 22:13:55 -07:00
Tara Ding	79825dbd32	Fix format	2026-04-27 22:11:47 -07:00
Tara Ding	139d4a7e86	fix format	2026-04-27 22:07:53 -07:00
Tara Ding	6251350cf4	Remove queue_time	2026-04-27 22:03:52 -07:00
Tara Ding	69f6272edc	Add benchmark for each node.	2026-04-27 21:39:57 -07:00
Matt Miller	ed201fff08	ci: dispatch tag push to Comfy-Org/cloud (#13541) ... Fires on v* tag push (earlier than release.published, which can lag) and triggers a repository_dispatch on Comfy-Org/cloud with event_type comfyui_tag_pushed. Legacy desktop dispatch in release-webhook.yml is left untouched.	2026-04-27 19:51:33 -07:00
rattus	b47f15f25a	fix: Handle un-inited meta-tensors in models (fixes a CPU TE crash) (CORE-67) (#13578)	2026-04-27 22:22:31 -04:00
Tara Ding	059b346966	fix server.py	2026-04-27 16:57:16 -07:00
Tara Ding	ca56e224a0	Moving collecting summary to benchmark_comfyui_serving	2026-04-27 16:50:40 -07:00
comfyanonymous	3cbf015578	Read audio and video at the same time in video loader node. (#13591)	2026-04-27 16:44:12 -07:00
Tara Ding	512deb3cd6	Fix returned vbench image filenames	2026-04-27 15:08:49 -07:00
Tara Ding	09f03107c2	remove checking existing png under input since example.png is always under input folder	2026-04-27 14:34:31 -07:00
Tara Ding	08411a1d65	Fix input dir	2026-04-27 14:28:46 -07:00
Tara Ding	125ed0be4b	Fix comments	2026-04-27 14:22:27 -07:00
Tara Ding	d407d82350	Fix comments	2026-04-27 14:20:32 -07:00
Tara Ding	ff5e379cc2	convert these two steps commands into one command: 1. check if downloading image or model is already there, if it is, skip. 2. remove prompt-file arg, when generating a new request, roundrobin the generated prompts	2026-04-27 14:13:01 -07:00
Tara Ding	52da6933b4	Add workflow	2026-04-27 14:01:06 -07:00
Tara Ding	978b962300	fix scripts	2026-04-27 13:42:53 -07:00
comfyanonymous	64b8457f55	ComfyUI v0.20.1 because github is broken again and messed up my release.	2026-04-27 16:10:14 -04:00
Tara Ding	8136fbbb4a	Fix input	2026-04-27 12:23:24 -07:00
Tara Ding	28bbdb0031	Fix vbench download	2026-04-27 11:53:35 -07:00
Tara Ding	c02b5d4c1e	Generate prompt file automatically.	2026-04-27 11:50:09 -07:00
comfyanonymous	75143eeb06	ComfyUI v0.20.0	2026-04-27 13:24:36 -04:00
Daxiong (Lin)	1233f077b1	chore: update workflow templates to v0.9.63 (#13586) ... Co-authored-by: Jedrzej Kosinski <kosinkadink1@gmail.com>	2026-04-27 10:06:03 -07:00
Alexander Piskun	6968a70e60	[Partner Nodes] HappyHorse model (#13582) ... * feat(api-nodes): add nodes for HappyHorse model Signed-off-by: bigcat88 <bigcat88@icloud.com> * fix price badges Signed-off-by: bigcat88 <bigcat88@icloud.com> * fix: allow durations up to 15 s Signed-off-by: bigcat88 <bigcat88@icloud.com> --------- Signed-off-by: bigcat88 <bigcat88@icloud.com>	2026-04-27 09:53:08 -07:00
comfyanonymous	115f418b64	Make EmptySD3LatentImage node use intermediate dtype. (#13577)	2026-04-26 23:23:57 -04:00
Tara Ding	00379b4acf	Move benchmark serving client into benchmarks folder	2026-04-26 19:41:55 -07:00
Tara Ding	96363fa74a	Revert "Add benchmark for model serving" ... This reverts commit ac85d7887f.	2026-04-26 16:48:58 -07:00
Daxiong (Lin)	7385eb2800	Add new ComfyUI blueprints and fix subgraph naming (#13371) ... * Remove local tag from subgraph name * New Subgraph blueprints * Remove duplicate blueprint * Update Subgraph size * Update subgraph * Update Blueprint * Remove local tag from subgraph name * New Subgraph blueprints * Remove duplicate blueprint * Update Subgraph size * Update subgraph * Update Blueprint * Update LTX 2.0 Pose to Video * Fix crop blueprint split coverage Made-with: Cursor * Clean up image edit blueprint metadata Made-with: Cursor * Update subgraph blueprints --------- Co-authored-by: Jedrzej Kosinski <kosinkadink1@gmail.com>	2026-04-26 22:59:16 +08:00
comfyanonymous	df22bcd5e1	Support loading the alpha channel of videos. (#13564) ... Not exposed in nodes yet.	2026-04-25 21:02:58 -04:00