Practical overview
When it fits
Use Uzu when building an application that needs to load an AI model and call inference through an integrated engine API.
Who it is for
- Developers integrating AI model inference into their own applications.
- Teams working in Rust, Python, TypeScript or Swift that want a documented engine and chat API.
Documented uses
- Add model inference to an application through a language binding.
- Look up and download a model, start a chat session, and request a reply.
- Evaluate an engine with model configuration and traceable computation features described by the project.
Requirements & limits
The repository identifies Uzu as open source under the MIT license.
- The repository documents four language routes: Rust, Python, TypeScript and Swift.
- The quick starts include model lookup and download before opening a chat session.
- The README describes unified model configurations and traceable computations.
- The project highlights unified memory on Apple devices, but the supplied material does not give a full compatibility matrix.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Set up a Python chat inference flow
Add the documented Uzu dependency and structure a Python program to load a model and request a chat reply.
Before you start
- Python environment
- uv package manager
- Uzu version 0.6.3
- Async Python code
Steps
Add Uzu with the command below.
uv add uzu==0.6.3Create an engine using
EngineConfig.create()andawait Engine.create(...).Look up the documented model identifier and download it, iterating over download progress.
Open a chat session, send system and user messages, and read the final reply.
Expected outcome
A Python application flow that obtains a model, starts a chat session and requests a text reply.
Guide limitations
- The example returns without continuing if the model lookup yields no model.
Set up a TypeScript chat inference flow
Add the documented TypeScript package and use the repository's chat quick-start pattern.
Before you start
- TypeScript project
- pnpm package manager
- Uzu package version 0.6.3
Steps
Add the package with
pnpm add @trymirai/uzu@0.6.3.Use the documented engine, model, download and chat-session pattern.
Submit system and user messages and retrieve the model reply.
Expected outcome
A TypeScript application flow structured to request a chat reply through Uzu.
Guide limitations
- The available README content truncates the TypeScript code sample, so it does not provide the complete runnable example.
Compare options for Model inference and serving
These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.
| Project | Documented purpose | Platform | Access and requirements |
|---|---|---|---|
| Uzu | Add model inference to an application through a language binding. | Linux, macOS, Server | The repository identifies Uzu as open source under the MIT license. |
| Morphingdb | Import a libtorch model and create predictions from values in a PostgreSQL table. | Docker, Linux, Self-hosted, Server | The README documents a Docker image and a build-from-repository option, followed by connecting to PostgreSQL and enabling the pgdl extension. |
| Gateway — pyadav | Send chat-completion requests to a provider through a common server endpoint. | Server | The repository is public and identifies the project as Apache-2.0 licensed. README installation instructions include binaries, platform package managers and Docker. |
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 11 October 2026
- Original website title
- trymirai/uzu · Official homepage title · retained in its source language