> ## Documentation Index
> Fetch the complete documentation index at: https://docs.muna.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# muna compile

> Compile a Python function for deployment.

The `compile` command compiles a Python function into a self-contained binary predictor that runs natively
on server, desktop, mobile, and web.

## Usage

```bash icon="terminal" theme={null}
$ muna compile [OPTIONS] PATH
```

| Argument | Description                     |
| :------- | :------------------------------ |
| `PATH`   | Path to your Python source file |

## Options

| Option        | Description                                                      |
| :------------ | :--------------------------------------------------------------- |
| `--overwrite` | Delete any existing predictor with the same tag before compiling |

## Compiling a Function

First, define a Python function with the `@compile` decorator:

```py greeting.py icon="python" theme={null}
from muna import compile

@compile(
    tag="@your-username/greeting",
    description="Say a friendly greeting."
)
def greeting(name: str) -> str:
    return f"Hey there {name}! We're glad you're using Muna."
```

Then compile it:

```bash icon="terminal" theme={null}
# 🔥 Compile the predictor
$ muna compile greeting.py
```

<Frame caption="Creating a predictor with the Muna CLI. This is realtime--not sped up.">
  <img src="https://www.muna.ai/compile.gif" />
</Frame>

## Overwriting Predictors

If you've already compiled a predictor with the same tag, use `--overwrite` to replace it:

```bash icon="terminal" theme={null}
# 🔥 Recompile and overwrite existing predictor
$ muna compile --overwrite greeting.py
```

<Warning>
  Overwriting a predictor immediately replaces the previous version. Any applications using the
  predictor will start using the new version on their next prediction.
</Warning>

## What Happens During Compilation

When you run `muna compile`, the CLI:

1. Parses your Python source file
2. Identifies functions decorated with `@compile`
3. Uploads the source code to Muna
4. Performs code generation and transpilation to native C++
5. Compiles binaries for all supported platforms

<Note>
  Compilation time varies based on function complexity. Check the status at
  [muna.ai/predictors](https://muna.ai/predictors).
</Note>

## Next Steps

<CardGroup cols={2}>
  <Card title="Defining Functions" icon="python" href="/predictors/create">
    Learn how to define compilable Python functions.
  </Card>

  <Card title="Function Requirements" icon="list-check" href="/predictors/requirements">
    Understand the requirements for compiled functions.
  </Card>
</CardGroup>