Local LLM runtime
MLX LM
MLX LM is a MIT-licensed local-LLM server-capable runtime for macOS, running on Metal, with an OpenAI-compatible API. Ease of use 4/5 in our sourced ratings.
- License
- MIT
- Operating systems
- macOS
- Backends
- Metal
- Install
- pip
- Ease of use
- 4/5
- Latest version
- v0.31.3
- GitHub stars
- 6,236
- Engine
- mlx
- Graphical app (GUI)
- Server mode
- OpenAI-compatible API
- CPU offload
- Multi-GPU
- Speculative decoding
- LoRA support
- KV-cache quantization
As of 2026-07-08 · source · GitHub
Quant-format support
What the feasibility gate checks before any VRAM math — per format and backend.
| Format | Backend | Supported |
|---|---|---|
| awq | Metal | no |
| mlx | Metal | yes |
Will your machine run a model with MLX LM?
Check with MLX LM preselected →Support facts come from each runtime's own documentation, stamped and dated — how the math works.