Pinned Loading

1. dflash-mlx dflash-mlx Public

   Lossless DFlash speculative decoding for MLX on Apple Silicon

   Python 273 14

2. mola mola Public

   Multi-LoRA inference server for Apple Silicon -- one base model, many adapters, zero reload

   Python 11

3. siliconrt siliconrt Public

   Experimental Apple Silicon inference substrate in C++ and Metal

   C++ 1