Rapid-MLX

View on GitHub

Rapid-MLX is an open-source (Apache 2.0) OpenAI- and Anthropic-compatible LLM inference server and Mac app for Apple Silicon, built on MLX, focused on reliable tool calling for coding agents. Release-gated with Claude Code, Codex CLI, Aider, Hermes and DeepSeek Harness.

★ 3.9K430 forksPythonCustomraullenchai

Local LLM inference server and macOS app for Apple Silicon, built on MLX. Provides OpenAI- and Anthropic-compatible APIs, continuous batching, and tool-calling support for coding agents.

Use Cases

Serve local LLMs on Apple SiliconProvide a local inference backend for coding agentsRun concurrent LLM requests with continuous batchingUse OpenAI- or Anthropic-compatible APIs with local modelsTest tool calling across agent clients

Built With

Language: Python
Frameworks: MLX · mlx-lm

Rapid-MLX

Use Cases

Built With

Tags