Commit Graph
5930 Commits
Author SHA1 Message Date
jaeone94 8754369880 Add native model and text encoder LoRA stack nodes 2026-09-14 12:31:39 +09:00
jaeone94 47e1363681 Reject non-integer DynamicGroup limits 2026-09-11 19:00:34 +09:00
jaeone94 df50b99091 Fix DynamicGroup input contract validation 2026-09-11 18:54:23 +09:00
jaeone94 a2cb2f84e9 Update DynamicGroup contract documentation 2026-09-11 17:55:46 +09:00
jaeone94andTalmaj Marinc 9d5ae2f68c Add DynamicGroup widget input
Validate submitted row counts independently of template field requirements,
preserve sparse positions within a fixed index limit, and keep empty groups
as lists in lazy callbacks.

Adapted from https://github.com/Comfy-Org/ComfyUI/pull/14507.

Co-authored-by: Talmaj Marinc <talmaj@comfy.org>
2026-09-11 14:16:58 +09:00
comfyanonymous 1d48d9cf7b Add linear to ImageColorSpace. (#16247) 2026-09-10 21:52:52 -04:00
comfyanonymous 6338e4bd42 Fix h3 fun controlnet with comfy compiler. (#16240) 2026-09-10 17:10:36 -04:00
Daxiong (Lin) 1f641fd933 chore: update workflow templates to v0.11.59 (#16233) 2026-09-11 01:39:20 +08:00
Alexander Piskun 5774ab9cc3 [Partner Nodes] feat(OpenRouter): add auto aspect ratio to the MAI image node (#16229)
Signed-off-by: Alexander Piskun <bigcat88@icloud.com>
2026-09-10 21:13:31 +04:00
Alexander Piskun 6eba895f7d [Partner Nodes] feat(Tripo): migrate to the v3 API, add the Smart Segment node and retire the dead widgets (#16201)
* [Partner Nodes] feat(Tripo): migrate all nodes to the v3 API and add the Smart Segment node

Signed-off-by: bigcat88 <bigcat88@icloud.com>

* [Partner Nodes] feat(Tripo): deprecate Text to Model, Image to Model and Texture model in favour of V2 nodes without the retired widgets

Signed-off-by: bigcat88 <bigcat88@icloud.com>

* [Partner Nodes] fix(Tripo): report the mixamo retarget limitation for rigs whose v3 task echo carries no model version

Signed-off-by: bigcat88 <bigcat88@icloud.com>

* [Partner Nodes] fix(Tripo): keep the Smart Segment widgets inside the source options so the frontend can build the node

Signed-off-by: bigcat88 <bigcat88@icloud.com>

* [Partner Nodes] fix(Tripo): correct the Retarget and Smart Segment tooltips to what Tripo accepts

Signed-off-by: bigcat88 <bigcat88@icloud.com>

---------

Signed-off-by: bigcat88 <bigcat88@icloud.com>
2026-09-10 20:55:49 +04:00
comfyanonymous a7b1d39d34 Update instructions for manual install on windows AMD. (#16217) 2026-09-09 22:33:59 -04:00
Alexander PiskunandPurz 7ea14e59da [Partner Nodes] feat(Gemini-LLM): Add Gemini 3.8 Flash to the Gemini text node (#16195)
Signed-off-by: Alexander Piskun <bigcat88@icloud.com>
Co-authored-by: Purz <97489706+purzbeats@users.noreply.github.com>
2026-09-09 20:14:27 -04:00
DELUXA be92396834 quant_ops: drop the dead ROCm triton arch gate (#16211) 2026-09-09 16:14:27 -04:00
Alexander PiskunandDaxiong 4989cdd954 [Partner Noes] feat(OpenRouter): add Microsoft mai-image-2.6 models (#16188)
* [Partner Nodes] feat(OpenRouter): add MAI-Image-2.6 image node

Signed-off-by: Alexander Piskun <bigcat88@icloud.com>

* [Partner Nodes] fix(api-client): show provider error details

Signed-off-by: Alexander Piskun <bigcat88@icloud.com>

---------

Signed-off-by: Alexander Piskun <bigcat88@icloud.com>
Co-authored-by: Daxiong (Lin) <contact@comfyui-wiki.com>
2026-09-10 02:23:21 +08:00
rattus 6f3895ed8d main: bump the AMD Windows VA quota to 4TB (#16199)
AMD+windows sets an unusually low maximum GPU virtual memory quota
compared to other plaforms. For reference, the hardware limits on
XT9060 are 128TB. On Nvidia RTX3060 it is 1TB. On nvidia RTX5090
it is 128TB.

Bump this into the same league, but keep it to a modest 4TB pending
further usage case need + testing going even higher to the hardware
limits.
2026-09-09 08:51:34 -07:00
Daxiong (Lin) 54e03f5367 Add new model blueprints and reorganize subgraph categories (#14785) 2026-09-08 22:08:46 -07:00
comfyanonymous 40c4fcdf51 ComfyUI v0.35.0 v0.35.0 2026-09-09 01:02:46 -04:00
Terry Jia 3074d0e331 Report the file saved by Save 3D (Advanced) as a standard 3d output item (#16171) 2026-09-08 21:50:21 -07:00
Daxiong (Lin) 6517734ff3 chore: update workflow templates to v0.11.57 (#16192) 2026-09-08 21:47:21 -07:00
Alexis Rolland 02dfb63b80 chores: Update tooltips of 3D nodes (CORE-323) (#16179) 2026-09-08 21:44:54 -07:00
Alexander Piskun be47aa22c0 [Partner Nodes] feat(OpenAI): add GPT Image 2.5 Flare and Sunburst models (#16190) 2026-09-08 21:41:31 -07:00
comfyanonymous 672ba9e5e3 Only lock repo PRs after merging if they contain a CLA signature. (#16191) 2026-09-08 20:42:15 -04:00
poorpaper 421a1c245c Fix MiniMax H3 denoise masks (#15988) 2026-09-08 14:25:33 -07:00
Alex Artyomov 0d0b6b5694 Add LTXVAddLatentGuide for pinning a pre-encoded latent as a guide (#16176) 2026-09-08 14:12:47 -07:00
comfyanonymous 249c5a3b95 Cleaner way of enabling quantized mm on text gen but not on text enc. (#16189) 2026-09-08 15:52:21 -04:00
Jukka Seppänen b7ebfd73c5 Fix quantized text encoder matmul gating and Gemma4 prefill cache positions (#16185) 2026-09-08 14:57:44 -04:00
Jukka Seppänen 488e8f8ab8 feat: Pixal3D multiview model support (CORE-421) (#16048) 2026-09-08 11:18:45 -07:00
rattus 00d34d92fe Comfy Aimdo 0.5.3 + Memory compiler fixes (#16180) 2026-09-08 12:32:43 -04:00
Adam Oster efa6c8f804 Add LTXV generated-keyframe nodes and Freeze Latent (#16040) 2026-09-07 19:02:58 -07:00
Alexis Rolland 5bbdf8a766 chore: Harmonize model attention nodes (#16154) 2026-09-07 20:05:59 -04:00
comfyanonymous f5ed117b88 Remove useless code. (#16169) 2026-09-07 19:27:58 -04:00
DELUXA 41db8f4fa1 Verify aotriton kernels actually launch before enabling pytorch attention (#15648) 2026-09-07 17:54:46 -04:00
guill 9ac7352f70 Fix registration issues (#15890) 2026-09-07 16:50:44 -04:00
Jialong(Bruce) Li 313a76fb8d Disable int8 weight-only quantization on devices without torch._int_mm (fixes MPS crash) (#16130) 2026-09-07 16:41:32 -04:00
Comfy Org PR Bot eb35786259 Bump comfyui-frontend-package to 1.51.10 (#16118) 2026-09-07 08:16:48 -07:00
Zhewen Tan ea33b15489 Fix transparent Porter-Duff blend modes (#15721) 2026-09-07 00:53:13 -07:00
Christian Byrne fbed745c8d Don't invert the alpha channel in the Invert Image node (#15622) 2026-09-06 19:38:26 -07:00
Christian Byrne f9c706f317 Don't blend the alpha channel in the Blend Images node (#15625) 2026-09-06 19:32:25 -07:00
Christian Byrne 20f1a41289 Fix Draw Text Overlay on images with an alpha channel (#15634) 2026-09-06 19:26:05 -07:00
Christian Byrne aa4582f93a Fix Quantize Image on images with an alpha channel (#15630) 2026-09-06 19:15:34 -07:00
Christian Byrne 25dfc16f9a Don't adjust the alpha channel in the image color adjustment nodes (#15629) 2026-09-06 18:09:20 -07:00
Christian Byrne d03a2430e1 Fix Detect Edges (Canny) on images with an alpha channel (#15631) 2026-09-06 14:13:35 -07:00
comfyanonymous 7b3b262bcb Add color space conversion node. (#16135)
Converts between sRGB, HDR (hlg) and HDR PQ.

Note that using this node will not make your images HDR, it will just put
the data as losslessly as it can in the other color space and only tonemap
when that isn't possible (ex: converting HDR to sRGB).
2026-09-06 17:02:02 -04:00
Christian Byrne 07dd46dc4a Fix the RGB/YUV conversion nodes on images with an alpha channel (#15632) 2026-09-06 13:59:43 -07:00
guill 82db4037ce Fix test_base_path_changes leaking a reloaded cli_args singleton (#16053) 2026-09-06 13:15:15 -07:00
rattus a99d1f9c14 Pause comfy compiler for long lived sparse allocations (#16148)
* model_prefetch: add friendly API to pause comfy compiler

* sparse_attention: Pause comfy compiler for un-freed pieces

The VSA plan and pool stats do not free within the transformer causing
a comfy-compiler rogue allocation and graph break.

Ideally these should be freed in the lifetime of the transformer run
(and comfy-compiler scope) but there isnt a convenient place for the
particular lifecylce intended by these variables. So go with the pause()
approach.
2026-09-06 10:49:45 -07:00
Daxiong (Lin)andAlexis Rolland 15eb748b3e chore: update embedded docs to v0.5.11 (#15947)
Co-authored-by: Alexis Rolland <alexisrolland@hotmail.com>
2026-09-06 09:31:05 +08:00
Jukka Seppänen e308cc73b4 Add Sparse Attention node (#16072)
* support VSA

* BlockSparseAttention node

* Add extra_tokens

* cleanup

* cleanup
2026-09-05 21:18:15 -04:00
comfyanonymous d1c5e64119 Fix potential mps issue of supports_fp64 called with None. (#16136) 2026-09-05 21:00:11 -04:00
comfyanonymous 18ebc2afd4 Update comfy-kitchen version to 0.2.33 (#16133) 2026-09-05 19:21:16 -04:00