fastmlx

FastMLX is a high performance production ready API to host MLX models.

OTHER License

Downloads
542
Stars
153

Bot releases are hidden (Show)

fastmlx - v0.2.1 Latest Release

Published by Blaizzy 2 months ago

What's Changed

Full Changelog: https://github.com/Blaizzy/fastmlx/compare/v0.2.0...v0.2.1

fastmlx - v0.2.0

Published by Blaizzy 2 months ago

What's New

  1. Environment Configuration (PR #15 by @SiddhantSadangi):

    • Set workers through environment variables
    • Improved default settings
  2. Tool Calling Support (PR #21 by @Blaizzy):

    • Added functionality for tool calling
    • Supports parallel tool calling mode
    • Models
      • Llama 3.1
      • Arcee Agent
      • Firefunction
      • xLAM
      • C4AI-Command-R-Plus
    • Modes
      • Standard (Without Streaming)
      • Parallel Tool Calling
  3. Refactor Vision Language Models stream_generate (PR #21 by @Blaizzy).

New Contributors

Full Changelog: https://github.com/Blaizzy/fastmlx/compare/v0.1.0...v0.2.0

fastmlx - v0.1.0

Published by Blaizzy 3 months ago

What's Changed

Fixes :

  • Cross origin Support #2
  • Max tokens not overriding #5

Full Changelog: https://github.com/Blaizzy/fastmlx/compare/v0.0.1...v0.1.0

fastmlx - v0.0.1

Published by Blaizzy 3 months ago

What's Changed

New Contributors

Full Changelog: https://github.com/Blaizzy/fastmlx/commits/v0.0.1