Skip to content

armgit / uzu

Public Imported from GitHub
0

A high-performance inference engine for AI models

3.8 KB
Raw Blame
---
title: 'About'
---

<Tip>
Looking for an instant start? Check out the drop-in snippets:

<Tabs>
    <Tab title="Python">
        <Columns cols={2}>
            <Card href="/quick-start/chat#python">
                Chat
            </Card>
            <Card href="/quick-start/chat-structured-output#python">
                Structured output
            </Card>
            <Card href="/quick-start/chat-cloud#python">
                Cloud
            </Card>
        </Columns>
    </Tab>
    <Tab title="Swift">
        <Columns cols={2}>
            <Card href="/quick-start/chat#swift">
                Chat
            </Card>
            <Card href="/quick-start/chat-structured-output#swift">
                Structured output
            </Card>
            <Card href="/quick-start/chat-cloud#swift">
                Cloud
            </Card>
        </Columns>
    </Tab>
    <Tab title="TypeScript">
        <Columns cols={2}>
            <Card href="/quick-start/chat#typescript">
                Chat
            </Card>
            <Card href="/quick-start/chat-structured-output#typescript">
                Structured output
            </Card>
            <Card href="/quick-start/chat-cloud#typescript">
                Cloud
            </Card>
        </Columns>
    </Tab>
    <Tab title="Rust">
        <Columns cols={2}>
            <Card href="/quick-start/chat#rust">
                Chat
            </Card>
            <Card href="/quick-start/chat-structured-output#rust">
                Structured output
            </Card>
            <Card href="/quick-start/chat-cloud#rust">
                Cloud
            </Card>
        </Columns>
    </Tab>
</Tabs>

</Tip>

[**Mirai**](https://trymirai.com/) lets you add high-performance AI right into your app with zero latency, full data privacy, and no inference costs. You don’t need an ML team or weeks of setup anymore. One developer can get it all running in minutes. To achieve this, we offer the following key products:

<Columns cols={3}>
    <Card title="uzu" img="/images/headers/uzu.jpg" href="https://github.com/trymirai/uzu">
        A Rust inference engine built to run AI with hardware specifics in mind.
    </Card>
    <Card title="lalamo" img="/images/headers/lalamo.jpg" href="https://github.com/trymirai/lalamo">
        A set of tools to optimize and convert models for on-device use.
    </Card>
    <Card title="cli" img="/images/headers/cli.jpg" href="https://github.com/trymirai/uzu#cli">
        A command-line tool to chat with models and serve them as a local API.
    </Card>
</Columns>

<Tip>
    If you run into a problem and don’t find an answer here, feel free to reach out in our [Discord](https://discord.com/invite/trymirai) community. We can even debug the issue with you over voice chat, or you can [book a call](https://calendar.google.com/calendar/u/0/appointments/schedules/AcZssZ0RkTVFVCqRHJdidCc0aRAr4T44KeO58N01tMrNBkjbNFndBAaTf6rKdtTTEryPo8JB_UNQ8Lx6) with our engineers ❤️.
</Tip>

## FAQ

<AccordionGroup>
    <Accordion title="Is this project open source?">
        <Check>
            The [**uzu**](https://github.com/trymirai/uzu) and [**lalamo**](https://github.com/trymirai/lalamo) libraries are fully open source under the MIT license.
        </Check>
    </Accordion>
    <Accordion title="Is it free to use?">
        <Check>
            The [**uzu**](https://github.com/trymirai/uzu) and [**lalamo**](https://github.com/trymirai/lalamo) libraries are completly free to use.
        </Check>
    </Accordion>
    <Accordion title="Which devices are supported?">
        <Info>Currently, only Apple Silicon (iOS/macOS) devices are supported.</Info>
    </Accordion>
    <Accordion title="Which models are supported?">
        <Info>The full list of supported models is available [here](https://trymirai.com/local-models).</Info>
    </Accordion>
</AccordionGroup>