Releases: comfyanonymous/ComfyUI
Releases · comfyanonymous/ComfyUI
v0.3.58
What's Changed
- Fix progress update crossover between users by @guill in #9706
- Add Hunyuan 3D 2.1 Support by @yousef-rafat in #8714
- Some changes to the previous hunyuan PR. by @comfyanonymous in #9725
- fix: add cache headers for images by @arjansingh in #9560
- Fix lowvram issues with hunyuan3d 2.1 by @comfyanonymous in #9735
- Print all fast options in --help by @comfyanonymous in #9737
- Enable bf16 VAE on RDNA4. by @comfyanonymous in #9746
- Don't enable pytorch attention on AMD if triton isn't available. by @comfyanonymous in #9747
- Fix amd_min_version crash when cpu device. by @comfyanonymous in #9754
- Add ZeroDivisionError catch for EasyCache logging statement by @Kosinkadink in #9768
- Add explicit casting in apply_rope for Qwen VL by @contentis in #9759
- Support qwen inpaint controlnet. by @comfyanonymous in #9772
- API nodes: add ByteDance video by @bigcat88 in #9712
- Small refactor of some vae code. by @comfyanonymous in #9787
- Change validate_inputs' output typehint to 'bool | str' and update docstrings by @Kosinkadink in #9786
- Fix issue on old torch. by @comfyanonymous in #9791
- Support hunyuan image 2.1 regular model. by @comfyanonymous in #9792
- Update template to 0.1.76 by @comfyui-wiki in #9793
- Fix lowvram issue with hunyuan image vae. by @comfyanonymous in #9794
- API nodes: add StabilityAI Audio API nodes by @bigcat88 in #9749
New Contributors
- @yousef-rafat made their first contribution in #8714
- @arjansingh made their first contribution in #9560
Full Changelog: v0.3.57...v0.3.58
v0.3.57
What's Changed
- convert Stable Cascade nodes to V3 schema by @bigcat88 in #9373
- [V3] convert Video nodes to V3 schema by @bigcat88 in #9489
- [V3] convert nodes_stability.py to V3 schema by @bigcat88 in #9497
- convert Primitive nodes to V3 schema by @bigcat88 in #9372
- SEEDS: update noise decomposition and refactor by @chaObserv in #9633
- Probably not necessary anymore. by @comfyanonymous in #9646
- Implement the USO subject identity lora. by @comfyanonymous in #9674
- Enable Convolution AutoTuning by @contentis in #9301
- USO style reference. by @comfyanonymous in #9677
- uso -> uxo/uno as requested. by @comfyanonymous in #9688
- Accept prompt_id in interrupt handler by @deepanjanroy in #9607
- ImageScaleToMaxDimension node. by @comfyanonymous in #9689
- Update template to 0.1.73 by @comfyui-wiki in #9686
- feat(api-nodes): add ByteDance Image nodes by @bigcat88 in #9477
- api nodes(Ideogram): add Ideogram Character by @bigcat88 in #9616
- [V3] convert Runway API nodes to the V3 schema by @bigcat88 in #9487
- Update comment in api example. by @comfyanonymous in #9708
- convert AlignYourStepsScheduler node to V3 schema by @bigcat88 in #9226
- Fix potential rope issue. by @comfyanonymous in #9710
- Update template to 0.1.75 by @comfyui-wiki in #9711
New Contributors
- @deepanjanroy made their first contribution in #9607
Full Changelog: v0.3.56...v0.3.57
v0.3.56
What's Changed
- Lower ram usage on windows. by @comfyanonymous in #9628
Full Changelog: v0.3.55...v0.3.56
v0.3.55
What's Changed
- WanSoundImageToVideoExtend node to manually extend s2v video. by @comfyanonymous in #9606
- Fix issue with s2v node when extending past audio length. by @comfyanonymous in #9608
- Add a LatentCut node to cut latents. by @comfyanonymous in #9609
- Support wan2.2 5B fun control model. by @comfyanonymous in #9611
- Support the Wan2.2 5B fun inpaint model. by @comfyanonymous in #9614
- Trim audio to video when saving video. by @comfyanonymous in #9617
- Update template to 0.1.70 by @comfyui-wiki in #9620
Full Changelog: v0.3.54...v0.3.55
v0.3.54
What's Changed
- Update nodes_model_patch.py by @comfyanonymous in #9597
Full Changelog: v0.3.53...v0.3.54
v0.3.53
What's Changed
- Fix EasyCache/LazyCache crash when tensor shape/dtype/device changes during sampling by @blepping in #9528
- Remove models from readme that are not fully implemented. by @comfyanonymous in #9535
- Add models/audio_encoders directory. by @comfyanonymous in #9548
- Implement wav2vec2 as an audio encoder model. by @comfyanonymous in #9549
- Make AudioEncoderOutput usable in v3 node schema. by @comfyanonymous in #9554
- Update template to 0.1.66 by @comfyui-wiki in #9557
- Adding Google Gemini Image API node by @Kosinkadink in #9566
- Update template to 0.1.68 by @comfyui-wiki in #9569
- WIP Wan 2.2 S2V model. by @comfyanonymous in #9568
- Negative audio in s2v should be zeros. by @comfyanonymous in #9578
- Fix #9537 by @KohakuBlueleaf in #9576
- Fixes to make controlnet type models work on qwen edit and kontext. by @comfyanonymous in #9581
- Improve s2v performance when generating videos longer than 120 frames. by @comfyanonymous in #9582
- Better s2v memory estimation. by @comfyanonymous in #9584
- Add DPM++ 2M SDE Heun (RES) sampler by @Balladie in #9542
- Add a LatentConcat node. by @comfyanonymous in #9587
- Adjust flux mem usage factor a bit. by @comfyanonymous in #9588
New Contributors
Full Changelog: v0.3.52...v0.3.53
v0.3.52
What's Changed
- LTXV: fix key frame noise mask dimensions for when real noise mask exists by @harelc in #9425
- Add that qwen edit model is supported to readme. by @comfyanonymous in #9463
- Support for Qwen Diffsynth Controlnets canny and depth. by @comfyanonymous in #9465
- Fix: Handle invalid filenames during workflow save (#9434) by @saurabh-pingale in #9445
- Forgot this. by @comfyanonymous in #9470
- Support diffsynth inpaint controlnet/model patch. by @comfyanonymous in #9471
- convert String nodes to V3 schema by @bigcat88 in #9370
- v3 nodes (part a) by @bigcat88 in #9149
- [V3] convert Google Veo API node to the V3 schema by @bigcat88 in #9272
- [V3] convert Ideogram API nodes to the V3 schema by @bigcat88 in #9278
- Make it easier to implement future qwen controlnets. by @comfyanonymous in #9485
- Support InstantX Qwen controlnet. by @comfyanonymous in #9488
- feat(api-nodes): change "OpenAI Chat" display name to "OpenAI ChatGPT" by @bigcat88 in #9443
- feat(api-nodes): add copy button to Gemini Chat node by @bigcat88 in #9440
- Update template to 0.1.65 by @comfyui-wiki in #9501
- Add elementwise fusions by @contentis in #9495
- Implement EasyCache and Invent LazyCache by @Kosinkadink in #9496
- Fix Conditioning masks on 3d latents. by @comfyanonymous in #9506
- Python 3.13 is well supported. by @comfyanonymous in #9511
- Don't use the annoying new navigation mode by default. by @comfyanonymous in #9518
- Update frontend to v1.25.10 and revert navigation mode override by @christian-byrne in #9522
New Contributors
- @saurabh-pingale made their first contribution in #9445
Full Changelog: v0.3.51...v0.3.52
v0.3.51 Qwen Image Edit model
What's Changed
- Put back frontend version. by @comfyanonymous in #9317
- Reduce portable size again. by @comfyanonymous in #9323
- SDPA backend priority by @contentis in #9299
- Make last PR not crash comfy on old pytorch. by @comfyanonymous in #9324
- Fix XPU iGPU regressions by @simonlui in #9322
- Added context window support to core sampling code by @Kosinkadink in #9238
- av is an essential dependency. by @comfyanonymous in #9341
- Make custom node testing checkbox optional in issue templates by @webfiltered in #9342
- Update CODEOWNERS by @yoland68 in #9343
- Update default parameters for Moonvalley video nodes by @guill in #9290
- convert WAN nodes to V3 schema by @bigcat88 in #9201
- Make SLG nodes work on Qwen Image model. by @comfyanonymous in #9345
- Avoid torch compile graphbreak for older pytorch versions by @StrongerXi in #9344
- Fix last commit not working on older pytorch. by @comfyanonymous in #9346
- Add warning when using old pytorch. by @comfyanonymous in #9347
- fix(OpenAIGPTImage1): set correct MIME type for multipart uploads by @bigcat88 in #9348
- Add FluxKontextMultiReferenceLatentMethod node. by @comfyanonymous in #9356
- Implement wan2.2 camera model. by @comfyanonymous in #9357
- Update frontend to v1.25.8 by @christian-byrne in #9361
- record audio node by @jtydhr88 in #8716
- Qwen image model refactor. by @comfyanonymous in #9375
- WIP Qwen edit model: The diffusion model part. by @comfyanonymous in #9383
- Update template to 0.1.60 by @comfyui-wiki in #9377
- Make context step index detection much more robust by @Kosinkadink in #9392
- Bump frontend to 1.25.9 by @christian-byrne in #9394
- fix(WAN-nodes): invalid nodeid for WanTrackToVideo by @bigcat88 in #9396
- P2 of qwen edit model. by @comfyanonymous in #9412
- api_nodes: add kling-v2-1 and v2-1-master by @bigcat88 in #9257
- api_nodes: add MinimaxHailuoVideoNode node by @bigcat88 in #9262
- api_nodes: add GPT-5 series models by @bigcat88 in #9325
- api_nodes: release version of google's models by @bigcat88 in #9304
- feat(api-nodes): add Vidu Video nodes by @bigcat88 in #9368
- Change the TextEncodeQwenImageEdit node to use logic closer to refererence code. by @comfyanonymous in #9432
- Rope fix for qwen vl. by @comfyanonymous in #9435
- Qwen rotary embeddings should now match reference code. by @comfyanonymous in #9437
- Disable prompt weights for qwen. by @comfyanonymous in #9438
- Fix bf16 precision issue with qwen image embeddings. by @comfyanonymous in #9441
- Bump template to 0.1.62 by @comfyui-wiki in #9419
New Contributors
- @StrongerXi made their first contribution in #9344
Full Changelog: v0.3.50...v0.3.51
v0.3.50
What's Changed
- Add Qwen Image model to readme. by @comfyanonymous in #9191
- Qwen Image model merging node. by @comfyanonymous in #9202
- qwenLora cannot load properly by @flybirdxx in #9208
- Update template to 0.1.52 by @comfyui-wiki in #9206
- Update frontend to v1.24.4 by @christian-byrne in #9175
- Fix RepeatLatentBatch not working on multi dim latents. by @comfyanonymous in #9227
- _ui.py import torchaudio safety check by @Kosinkadink in #9234
- async API nodes by @bigcat88 in #9129
- Users report gfx1201 is buggy on flux with pytorch attention. by @comfyanonymous in #9244
- Not sure if AMD actually support fp16 acc but it doesn't crash. by @comfyanonymous in #9258
- Bump pytorch cuda and rocm versions in readme instructions. by @comfyanonymous in #9273
- Only show feature flags log when verbose. by @comfyanonymous in #9281
- remove creation of non-used asyncio_loop by @bigcat88 in #9284
- Update template & embedded docs by @comfyui-wiki in #9283
- Support SimpleTuner lycoris lora for Qwen-Image by @PsychoLogicAu in #9280
- fix(Kling Image API Node): do not pass "image_type" when no image by @bigcat88 in #9271
- Update template to 0.1.58 by @comfyui-wiki in #9302
- Update test release package workflow with python 3.13 cu129. by @comfyanonymous in #9306
- Wan2.2 fun control support. by @comfyanonymous in #9292
- Make torchaudio exception catching less specific by @Kosinkadink in #9309
- Update template to 0.1.59 by @comfyui-wiki in #9313
- Update release workflow to python3.13 pytorch cu129 by @comfyanonymous in #9315
- Downgrade frontend for release. by @comfyanonymous in #9316
New Contributors
- @flybirdxx made their first contribution in #9208
Full Changelog: v0.3.49...v0.3.50
v0.3.49
What's Changed
- API for Recently Used Items by @Myestery in #8792
- Update template to 0.1.47 by @comfyui-wiki in #9153
- Make sure all the conds are on the right device. by @comfyanonymous in #9151
- Make sure context is on the right device. by @comfyanonymous in #9154
- Cleanup. by @comfyanonymous in #9160
- Lower cond vram use by casting at the same time as device transfer. by @comfyanonymous in #9159
- Fix broken controlnet from last PR. by @comfyanonymous in #9167
- Various fixes for broken things from earlier PR. by @comfyanonymous in #9168
- Add some warnings and prevent crash when cond devices don't match. by @comfyanonymous in #9169
- Fix another issue with the PR. by @comfyanonymous in #9170
- Initial support for qwen image model. by @comfyanonymous in #9179
- Add Veo3 video generation node with audio support by @christian-byrne in #9110
- Update template to 0.1.48 by @comfyui-wiki in #9182
- Fix default shift and any latent size for qwen image model. by @comfyanonymous in #9186
- Update template to 0.1.51 by @comfyui-wiki in #9187
New Contributors
Full Changelog: v0.3.48...v0.3.49