v0.10.0
📦 comfyuiView on GitHub →
✨ 5 features🐛 4 fixes🔧 8 symbols
Summary
This release updates workflow templates, introduces new features like extended ByteDance models and the Bria Edit node, and includes fixes for AMD GPU compatibility and Autogrow validation.
Migration Steps
- Adjust memory usage factor calculation for flux2 klein.
✨ New Features
- Extend ByteDance nodes with seedance-1-5-pro model.
- Add image sizes to clip vision outputs.
- Add advanced parameter to Input classes for advanced widgets support.
- Auto-discover all nodes_*.py files in API Nodes.
- Add Bria Edit node.
🐛 Bug Fixes
- Fixed TypeError on AMD GPUs by moving audio tensor to CPU before numpy conversion.
- Make Autogrow validation work properly.
- Fixed issue #11963.
- Convert mono audio to fake stereo for LTXV VAE encoding.