tabby/.changes/v0.13.1.md
2024-07-10 10:24:36 +09:00

7 lines
268 B
Markdown

## v0.13.1 (2024-07-10)
### Fixed and Improvements
* Bump llama.cpp version to b3334, supporting Deepseek V2 series models.
* Turn on fast attention for Qwen2-1.5B model to fix the quantization error.
* Properly set number of GPU layers (to zero) when device is CPU.