v1.9.0
gristlabs/grist-corev1.9.0Mar 1, 2026by FranckyB
AI Summary
Performance improvements focused on CUDA Graphs acceleration for Qwen3-TTS and Multi-GPU support for subsystems.
Key Highlights
- 5-10x faster inference via Faster-Qwen3-TTS integration
- Multi-GPU assignment for TTS, ASR, and Llama.cpp
- Automatic fallback to standard Qwen3TTSModel when CUDA graphs fail
New Features
- CUDA Graphs acceleration (CUDA only)
- GPU assignment dropdowns and persistence
- Trained model caching between generations
Full Release Notes
**Faster-Qwen3-TTS Integration** - **5-10x Faster Inference** - Integrated [Faster-Qwen3-TTS](https://github.com/andimarafioti/faster-qwen3-tts) for CUDA graph-accelerated Qwen3 generation with bit-identical output quality - **All Qwen3 Models** - Acceleration applies to Base, CustomVoice, VoiceDesign, and Trained Model checkpoints - **Toggle in Settings** - Enable/disable CUDA Graphs Acceleration under Faster-Qwen3-TTS section (CUDA only) - **Automatic Fallback** - Gracefully falls back to standard Qwen3TTSModel when CUDA graphs are unavailable or fail - **Setup Scripts Updated** - `setup-windows.bat` and `setup-linux.sh` auto-install the package (not on macOS) - **Trained Model Caching** - Trained model checkpoints are now cached between generations instead of reloading every time **Multi-GPU Support** - **GPU Assignment Dropdowns** - Assign TTS, ASR, and Llama.cpp to different GPUs on multi-GPU systems - **Per-Subsystem Control** - Each subsystem (TTS, ASR, LLM) can run on a separate GPU to maximize throughput - **Automatic Detection** - GPU dropdowns only appear when multiple CUDA GPUs are detected - **Saved Preferences** - GPU assignments persist across restarts via config.json **Bug Fixes** - **Conversation Tab Fix** - Fixed a broken component reference (`vv_conv_sentences_per_chunk` → `vv_conv_paragraph_per_chunk`) that prevented all interactive elements from working on the Conversation tab