Updated

MLX LM

Apple's open-source package for running and fine-tuning language models on Apple silicon with the MLX framework.

22 open-source projects declare MLX LM in their code; no maker's live product we track uses it yet.

Open sourceSelf-hostable

Profile checked

Ask your AI about this, with this page as the source:ChatGPT ↗Claude ↗Perplexity ↗

How makers use MLX LM

What MLX LM is used for

Run Models Locally · Inference engine →Running and fine-tuning models on Apple silicon Macs, using unified memory efficiently.Trade-off: Apple silicon only, and models need MLX-format weights.
Ollama39Llama21LM Studio4llama.cpp1Jan0MLX LM0Makers' products on each, for run models locally

MLX LM vs the alternatives makers chose

OptionMakers' productsOpen source
Ollama39 64%Yes
Llama21 34%No
LM Studio4 7%No
llama.cpp1 2%Yes
MLX LM0 0%Yes

For run models locally: how many of the 61 makers' products we know the run models locally of chose each option.

When to use MLX LM, and when not

Pick MLX LM if

  • You work on a Mac with Apple silicon and want the fastest local inference or fine-tuning there. Run Models Locally →

Something else fits better if

  • You want a local model behind an API for your code and dev tools in minutes: Ollama.
  • You'd rather browse and compare models in a desktop app before writing code: LM Studio or Jan.
  • You ship local inference inside your own app or need settings the apps don't expose: llama.cpp.
  • You need a model family with a permissive license for commercial use: Qwen or gpt-oss or Gemma.
From each decision's guide, written from what makers' products chose and why.

How MLX LM itself is built

1 tools, from its own code, website and Product Hunt page.

Who uses it

No maker's live product on record yet, and 22 open-source projects that declare it in their code.

We haven't found a maker's live product that uses MLX LM yet — only the open-source projects below, which declare it in their code.

self-llm
self-llm《开源大模型食用指南》针对中国宝宝量身打造的基于Linux环境快速微调(全参数/Lora)、部署国内外开源大模型(LLM)/多模态大模型(MLLM)教程Run Models LocallyIn its code · source ↗

Open source: a project that declares MLX LM as a dependency in its public code — verifiable, but not necessarily a live product.

Alternatives to MLX LM

All alternatives by situation →

Questions makers ask about MLX LM

What are the alternatives to MLX LM?

For run models locally, the other options makers' products we track chose most are Ollama (39), Llama (21), LM Studio (4), llama.cpp (1). Open source or self-hostable among them: Ollama, Llama, LM Studio, llama.cpp. Each alternative, with the situation it fits, is on the MLX LM alternatives page.

When should you use MLX LM?

You work on a Mac with Apple silicon and want the fastest local inference or fine-tuning there (run models locally). Something else fits better if: you want a local model behind an API for your code and dev tools in minutes, pick Ollama; you'd rather browse and compare models in a desktop app before writing code, pick LM Studio or Jan.

Is MLX LM open source or self-hostable?

Open source, and you can self-host it. source ↗

Can AI coding agents work with MLX LM?

No llms.txt, official MCP server or CLI found yet.

Who uses MLX LM?

No maker's live product we track yet; 22 open-source projects declare it in their code. source ↗