trymiraiOfficial·8 Agent Skills Included

uzu

Run AI models directly inside your app with zero latency

Runs AI model inference directly inside your app on the user's device, with no cloud servers or API costs. Eliminates network latency, data privacy risks, and per-request inference fees by keeping everything local. Provides simple Rust, Python, Swift, and TypeScript APIs that handle model downloading, configuration, and chat sessions automatically.

1,670|68

↗

npx skills add trymirai/uzu --all -g -y

All Skills in This Repository (8)

Pure Emerald Level Indicators

📦 In Repo

trymirai

validate-inference

Validates inference outputs by comparing them with reference results via Rust cargo test harness.

list-downloaded-models

List downloaded models by inspecting the ./models/{ENGINE_VERSION} directory structure.

architecture

Document the core architecture and component interactions of the uzu inference engine.

download-model

Download test models by REPO_ID into ./models/{ENGINE_VERSION}/{MODEL_NAME}.

trace-model

Validate per-layer AI model inference outputs against traces.safetensors.

metal-gpu-debug

Automate Metal GPU trace collection, parsing, and analysis on macOS.

list-models

List models supported by the inference engine via the Python helper.

benchmark-model

Benchmark AI model inference latency, throughput, and resource usage via CLI.

Official

Intermediate

Frequently Asked Questions

FAQPage Schema

How to install uzu?▼

Run `npx skills add trymirai/uzu --all -g -y` in your terminal to install all skills in this suite globally.

What does uzu do?▼

Uzu is a high-performance inference engine that lets you run AI models directly inside your app on the user's device, with zero latency, full data privacy, and no inference costs.

Which programming languages does uzu support?▼

Uzu provides ready-to-use APIs for Rust, Python, Swift, and TypeScript, so you can embed on-device AI in mobile, desktop, or server apps.

Does uzu work on iPhones and Macs?▼

Yes. Uzu is optimized for Apple devices and uses unified memory on iOS and macOS for fast, efficient local inference.

Do I need to manage model downloads manually?▼

No. Uzu handles model downloading, configuration, and chat session setup automatically through its high-level API.