mlx-serve
mlx-serve is a local LLM inference server for Apple Silicon, supporting OpenAI and Anthropic APIs.
- Region
- Overseas
- Pricing
- Free
- Open source
- Yes
- GitHub Stars
- ★ 1.7k
- Views
- 2
- Source
- GitHub
- Added
- 2026-09-30
- Last verified
- 2026-09-30
Overview
mlx-serve is a macOS application built with Zig and Swift that can run any LLM model on Apple Silicon devices. It provides OpenAI and Anthropic-compatible HTTP APIs, supporting text, image, video, music, and speech generation. Users can operate it via a menu bar app or command-line interface with no configuration required. The tool is ideal for developers and researchers, offering a fast and efficient local inference solution.
Key features
- ▪Supports MLX and GGUF model formats
- ▪OpenAI and Anthropic API compatible
- ▪Supports multi-modal content generation (text, image, video, music)
- ▪Built-in MCP tool calling
Use cases
Pros
- ＋High-performance local inference
- ＋Multiple API compatibility
- ＋Rich feature support
- ＋Easy installation and usage
Limitations / notes
- －Only supports macOS
- －Requires some technical background
Who it's for
This overview was compiled by AI from public sources and may contain inaccuracies — please refer to the official site.
FAQ
Is it free?
Yes, mlx-serve is free and open-source.
Does it support Chinese?
The README includes a simplified Chinese version, but specific support details need further confirmation.
Can it be used commercially?
mlx-serve is licensed under the MIT license and can be used for commercial purposes.
Similar agents
Cursor
FeaturedAI-powered code editor with Agent mode
Dify
FeaturedOpen-source LLMOps + Agent orchestration platform
扣子 Coze
FeaturedByteDance's no-code Agent/Bot building platform
Trae
FeaturedAI IDE (with Agent) from ByteDance
LangChain
FeaturedThe most mainstream LLM/Agent orchestration framework
Devin
FeaturedAutonomous software engineer Agent by Cognition
Something wrong? Let us know on the About page and we'll fix it.