mizorewww/laya-mlx
6,928 561PythonApache-2.0Updated 2 weeks ago
Native MLX runtime for Laya typed decision models — 7–14 ms short decisions on M3 Max. No text generation, PyTorch, or cloud API.
apple-silicondecision-modelinferencelayalocal-aimachine-learningmlxmodernbertsystem-onetyped-decisions
Our verdict: worth using
Provides a native MLX runtime for typed decision models with low latency, suitable for local inference on Apple Silicon.
Filed under llm serving in our directory.