trymirai/uzu
High-performance AI model inference engine
- Chart
- Not on a chart right now
- Stars
- 3.4k
What it is
A high-performance inference engine for deploying AI models directly inside applications. It handles model downloading and inference configuration automatically, with bindings for Rust, Python, Swift, and TypeScript.
Who it is for
Application developers who want to run AI models locally on-device, including Apple hardware, or via cloud APIs.
Why it is trending
It gained 1,013 stars today following the recent 0.7.1 release. Its unified API supports both local models utilizing Apple unified memory and cloud models.
Main features
- High-level APIs for Rust, Python, Swift, and TypeScript
- Automatic model downloading and inference configuration
- Unified memory utilization on Apple devices
- Shared chat session and instance reuse without reloading models
Summarized on November 18, 2026 from the README and repository data. It can be wrong; check the source before relying on it. Report a wrong summary