Release v0.3.5

Updated

• improved fastAPI server
• support libra confidence router

Release v0.3.4

Updated

• improved the hidden states generation method
• project structure refactoring

Release v0.3.3

Added

• langchain integration
• local_rag and graph_rag example

Updated

• generate method to support hidden states output

Release v0.3.2

Added

• model management, FastAPI-server
• unit test

Updated

• synchronized with the mlx-lm
• simplified README

Release v0.3.1

Updated

• updated mlx_fastchat_worker for supporting mlx >= 0.14.
• updated conda config.

Release v0.3.0

Added

• Lora support for GBA low-bit models.

Release v0.2.1

Added

• support for Phi-3

Release v0.2.0

Added

• Conversion: Utilize gba2mlx.py to convert models from GBA format to a format compatible with the MLX framework, ensuring smooth integration and optimal performance.
• Generation: Includes scripts for generating content using GBA quantized models within the MLX environment, empowering users to leverage the advanced capabilities of GBA models for natural language content creation.
• Fully support GreenBitAI's MLX Model Collection