v0.3.53
📦 comfyuiView on GitHub →
✨ 4 features🐛 6 fixes🔧 6 symbols
Summary
This release introduces new features like the Gemini Image API node and wav2vec2 audio encoder, alongside several bug fixes related to caching, S2V performance, and ControlNet compatibility.
✨ New Features
- Implement wav2vec2 as an audio encoder model.
- Adding Google Gemini Image API node.
- Add DPM++ 2M SDE Heun (RES) sampler.
- Add a LatentConcat node.
🐛 Bug Fixes
- Fix EasyCache/LazyCache crash when tensor shape/dtype/device changes during sampling.
- Make AudioEncoderOutput usable in v3 node schema.
- Negative audio in s2v should be zeros.
- Fixes to make controlnet type models work on qwen edit and kontext.
- Improve s2v performance when generating videos longer than 120 frames.
- Better s2v memory estimation.
🔧 Affected Symbols
EasyCacheLazyCacheAudioEncoderOutputcontrolnet type modelsqwen editkontext