trymirai/uzu

A high-performance inference engine for AI models

62
/ 100
Established

Leverages Apple Silicon's unified memory architecture with Metal-accelerated kernels for optimized on-device inference. Supports a custom model format with conversion tooling (via lalamo) for popular open-source models, and provides language bindings for Swift and TypeScript alongside a Rust core API. Includes built-in CLI utilities for model serving, benchmarking, and inference with configurable decoding parameters.

1,492 stars. Actively maintained with 69 commits in the last 30 days.

No Package No Dependents
Maintenance 25 / 25
Adoption 10 / 25
Maturity 15 / 25
Community 12 / 25

How are scores calculated?

Stars

1,492

Forks

44

Language

Rust

License

MIT

Last pushed

Mar 13, 2026

Commits (30d)

69

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/trymirai/uzu"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.