Rare find

Olleh. Use Apple's built-in AI models from your terminal, the same way you'd use other AI tools.

github.com/mattt/olleh

Vaya's read on this project

Problem, audience, market, and the verdict — sign in to see it.

Updates

September 2025
  • Release —1.1.1
  • Update FoundationModels usage for macOS 26 GM APIs
July 2025
  • Transfer project from Loopwork (#19)
  • Release —1.1.0
  • Add support for loading Foundation Models adapter (#18)
June 2025
  • Fix CI workflow
  • Add logo (#16)
  • Update installation instructions
  • Release —1.0.0
  • Update README
  • Overhaul model information functionality (#15)
  • Implement comprehensive server logging (#14)
  • Add CI workflow (#13)
  • Ignore empty lines entered by user (#12)
  • Refactor spinner animation (#11)
  • Update default port to `11941` (#10)
  • Improve `olleh run` UX (#9)
  • Add missing fields to server responses (#8)
  • Fix timestamp formatting for server responses (#7)
  • Add missing stream writer finalizer (#6)
  • Add chat generation parameters and streaming (#5)
  • Fix how configured parameters are passed as generation options (#4)
  • Optimize language model generation (#3)