OpenNavo

Ollama CLI

Ollama · The Ollama command-line distribution for running open models, managing local inference and exposing a REST API; this cask does not install the desktop app.

CaskAI ToolsApple silicon · Intel
Homepage
From homebrew/cask
30-day installs185#524 in App
1-year installs185Last 365 days
Latest version0.40.110 hours ago · Oct 7
Download size167MBarm64 / x86_64 · tgz
Release cadenceWeekly8 releases in 30 days

  1. 0.40.1

    LatestPatchOct 7 · Wed

    OpenNavo editorialAdded to Homebrew Oct 8 05:37

    Removes the account step from CLI onboarding, proxies cloud usage and balance APIs, and fixes Windows Clef reads and manifest symlinks.

    Changes
    Removes the account step from CLI onboarding.Proxies cloud usage and balance APIs through the server.Fixes Clef head reads beyond 2 GiB on Windows.Avoids manifest symlinks on Windows.

  2. 0.40.0

    Sep 25 · Fri

    OpenNavo editorialAdded to Homebrew Oct 6 19:26

    Automatically runs supported model architectures on MLX on Apple Silicon, including chat and decision models, and adds an MLX embedding model.

    Changes
    Uses MLX by default for supported architectures on Apple Silicon.Supports decision models on MLX.Adds embeddinggemma-2 support on MLX.

  3. 0.35.1

    PatchSep 29 · Tue

    OpenNavo editorialAdded to Homebrew Oct 2 20:20

    Adds multimodal Clef decision models, raises web searches per response to ten, and preserves Modelfile capability declarations; updates inference engines.

    Changes
    Adds multimodal Clef and Clef Flash through /v1/systemone.Raises web searches per response from three to ten.Preserves explicit Modelfile CAPABILITY declarations.Reports decision-only capabilities for decision models.Updates llama.cpp and the MLX engine.

An open-source interface for Homebrew. Installs run through the macOS app or the brew command.

© 2026 OpenNavo