{
    "product": {
        "slug": "mlx-lm",
        "name": "MLX LM",
        "kind": "app",
        "website": "https://github.com/ml-explore/mlx-lm",
        "page": "https://listme.name/software/mlx-lm",
        "tagline": "Python package for running and tuning language models on Mac",
        "summary": "MLX LM is an MIT-licensed Python package for running and fine-tuning language models with the MLX framework on Mac computers. It installs with pip or conda, supports distributed inference, and the README says large models need macOS 15.0 or later.",
        "platforms": [
            "macos",
            "cli"
        ],
        "pricing": "open-source",
        "price_note": "MIT-licensed package is free to install and use",
        "free_to_use": true,
        "open_source": true,
        "license": "MIT",
        "source": "https://github.com/ml-explore/mlx-lm",
        "strengths": [
            "MIT-licensed Python package",
            "Installs with pip or conda",
            "Distributed inference with mx.distributed"
        ],
        "limits": [
            "Runs only on Mac computers",
            "Large models need macOS 15.0 or higher"
        ],
        "checked": "2026-10-10",
        "connected_to_publisher": false,
        "overview": "MLX LM provides command-line and Python tools for loading models in the MLX format and generating text with them. The README shows installation with pip or conda and a load-and-generate example. It also lists distributed inference and fine-tuning through mx.distributed. The repository is MIT licensed and had about 7,300 stars when checked.\n\nThe README says large models require macOS 15.0 or higher, so the package is limited to Mac computers. It suits Mac users who want to script local models in Python or tune them. Users on Windows or Linux may compare Ollama or llama.cpp, which are both in this ranking.",
        "highlights": [
            "Python package",
            "Command-line tools",
            "Fine-tuning",
            "Distributed inference",
            "Model loading and generation"
        ],
        "json": "https://listme.name/api/v1/products/mlx-lm.json"
    },
    "rankings": [
        {
            "list": "local-llm-runners",
            "name": "Local LLM runners",
            "page": "https://listme.name/best/local-llm-runners",
            "rank": 9,
            "of": 10,
            "score": 80,
            "verdict": "MIT-licensed Python package for running and tuning models with MLX on Mac, installed with pip or conda. Large models need macOS 15.0 or later, and the tool is limited to Mac computers, so it ranks below the cross-platform runners.",
            "best_for": "Python scripting of local models on a Mac"
        }
    ],
    "alternatives": [
        {
            "slug": "ollama",
            "name": "Ollama",
            "score": 92,
            "page": "https://listme.name/software/ollama"
        },
        {
            "slug": "llama-cpp",
            "name": "llama.cpp",
            "score": 90,
            "page": "https://listme.name/software/llama-cpp"
        },
        {
            "slug": "lm-studio",
            "name": "LM Studio",
            "score": 89,
            "page": "https://listme.name/software/lm-studio"
        },
        {
            "slug": "gpt4all",
            "name": "GPT4All",
            "score": 87.5,
            "page": "https://listme.name/software/gpt4all"
        },
        {
            "slug": "jan",
            "name": "Jan",
            "score": 86,
            "page": "https://listme.name/software/jan"
        },
        {
            "slug": "koboldcpp",
            "name": "KoboldCpp",
            "score": 84.5,
            "page": "https://listme.name/software/koboldcpp"
        },
        {
            "slug": "textgen",
            "name": "textgen",
            "score": 82.5,
            "page": "https://listme.name/software/textgen"
        },
        {
            "slug": "llamafile",
            "name": "llamafile",
            "score": 81,
            "page": "https://listme.name/software/llamafile"
        },
        {
            "slug": "localai",
            "name": "LocalAI",
            "score": 79,
            "page": "https://listme.name/software/localai"
        }
    ],
    "links": {
        "page": "https://listme.name/software/mlx-lm",
        "markdown": "https://listme.name/software/mlx-lm.md",
        "badge": "https://listme.name/badge/mlx-lm.svg",
        "alternatives_page": "https://listme.name/alternatives/mlx-lm",
        "alternatives_json": "https://listme.name/api/v1/alternatives/mlx-lm.json"
    },
    "source": "https://listme.name",
    "license": "Free to read and cite with attribution to listme.name.",
    "docs": "https://listme.name/developers"
}