🍺 BREW Explorer

← all formulae

mlx

brew install mlx v0.31.2 MIT AND Apache-2.0

Array computing framework optimized for Apple silicon with NumPy-like Python API.

Why you might care

MLX is built from the ground up for Metal on Apple silicon, offering fast tensor operations and machine learning workflows without cloud infrastructure. It targets researchers and ML engineers deploying on Macs, providing a low-level C++ core with Python bindings that integrate naturally into existing NumPy-based workflows.

Categories

data-science ml library

Alternatives

NumPy JAX PyTorch TensorFlow
8.4k
30-day installs · #400
29.9k
90-day · #397
58.5k
365-day · #605

Runtime dependencies

python@3.14

Build dependencies

cmake fmt nanobind nlohmann-json python-setuptools robin-map

Links

Blurb generated by claude-haiku-4-5 on 2mo ago.

Raw metadata 
{
  "aliases": [],
  "alternatives": [
    "NumPy",
    "JAX",
    "PyTorch",
    "TensorFlow"
  ],
  "build_dependencies": [
    "cmake",
    "fmt",
    "nanobind",
    "nlohmann-json",
    "python-setuptools",
    "robin-map"
  ],
  "categories": [
    "data-science",
    "ml",
    "library"
  ],
  "caveats": null,
  "conflicts_with": [],
  "dependencies": [
    "python@3.14"
  ],
  "deprecated": 0,
  "deprecation_reason": null,
  "desc": "Array framework for Apple silicon",
  "disable_reason": null,
  "disabled": 0,
  "enrichment_fetched_at": "2026-06-20T23:40:50+00:00",
  "first_seen": "2026-06-20T23:34:18+00:00",
  "full_name": "mlx",
  "github_default_branch": null,
  "github_last_commit_at": null,
  "github_readme_excerpt": null,
  "github_repo": null,
  "github_stars": null,
  "github_topics": [],
  "homepage": "https://ml-explore.github.io/mlx/build/html/index.html",
  "homepage_og_description": null,
  "homepage_og_image": null,
  "homepage_title": "MLX \u2014 MLX 0.31.2 documentation",
  "installs_30d": 8414,
  "installs_365d": 58496,
  "installs_90d": 29894,
  "keg_only": 0,
  "keg_only_reason": null,
  "last_seen": "2026-06-20T23:34:18+00:00",
  "license": "MIT AND Apache-2.0",
  "llm_generated_at": "2026-06-20T23:44:37+00:00",
  "llm_model": "claude-haiku-4-5",
  "name": "mlx",
  "oldnames": [],
  "one_liner": "Array computing framework optimized for Apple silicon with NumPy-like Python API.",
  "optional_dependencies": [],
  "rank_30d": 400,
  "rank_365d": 605,
  "rank_90d": 397,
  "raw_hash": "f52f902e83515b21",
  "recommended_dependencies": [],
  "revision": 0,
  "ruby_source_path": "Formula/m/mlx.rb",
  "tap": "homebrew/core",
  "test_dependencies": [],
  "uses_from_macos": [],
  "version_head": "HEAD",
  "version_stable": "0.31.2",
  "versioned_formulae": [],
  "why_use_this": "MLX is built from the ground up for Metal on Apple silicon, offering fast tensor operations and machine learning workflows without cloud infrastructure. It targets researchers and ML engineers deploying on Macs, providing a low-level C++ core with Python bindings that integrate naturally into existing NumPy-based workflows."
}