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

# Installation

> Install conformalforecast, verify it imports, and pick the optional extras you need.

## From PyPI

```bash theme={null}
pip install conformalforecast
```

Or with uv:

```bash theme={null}
uv add conformalforecast
```

The core install is pure Python and depends only on `narwhals`, `numpy`,
`pydantic` and `scipy`; pandas and polars are both optional. Python 3.10
or newer is required.

## Verify the install

```bash theme={null}
python -c "import conformalforecast; print(conformalforecast.__version__)"
```

## Optional extras

Everything beyond the calibration math is an extra, so the base install
never pulls a forecasting framework:

| Extra | Installs | Needed for |
| - | - | - |
| `nixtla` | statsforecast, mlforecast, neuralforecast, utilsforecast | the Nixtla adapters |
| `spci` | quantile-forest | `SequentialPredictiveConformalInference` |
| `plotting` | matplotlib | plotting helpers |
| `all` | all of the above | — |

```bash theme={null}
pip install "conformalforecast[nixtla]"
```

> **Importing the package never imports a forecasting library**
>
> The Nixtla adapters are re-exported lazily, so
> `import conformalforecast` does not import statsforecast, mlforecast,
> neuralforecast or torch. The library is imported the first time you
> touch the adapter that needs it, and if the extra is missing you get
> an `ImportError` naming the exact `pip install` command.

`CallableAdapter` needs no extra at all — it wraps any function and is
pure numpy, which makes it the quickest way to try the package against a
model you already have.

## From a git checkout

For development, clone the repository and use the `devenv` target, which
syncs every extra plus the dev group and installs the pre-commit hooks:

```bash theme={null}
git clone https://github.com/Nixtla/conformalforecast.git
cd conformalforecast
make devenv
```

See
[CONTRIBUTING](https://github.com/Nixtla/conformalforecast/blob/main/CONTRIBUTING.md)
for the test, lint and documentation workflow.

> **Alpha versioning**
>
> `conformalforecast` is in early development and the API is unstable
> until v0.2.0. Pin a version if you depend on exact interval values.

## Next

* [Quick start](quickstart) — calibrated intervals in one screen.
* [Methods](../methods/index) — the seven methods, and how to choose
  one.
* [Nixtlaverse](../adapters/index) — statsforecast, mlforecast,
  neuralforecast.
* [Architecture](../concepts/architecture) — one class per method, its
  two surfaces, and which one you want.


## Related topics

- [Installation](/neuralforecast/docs/getting-started/installation.html.md)
- [Statistical ⚡️ Forecast](/statsforecast/index.html.md)
- [Hierarchical Forecast 👑](/hierarchicalforecast/index.html.md)
- [Ray](/statsforecast/docs/distributed/ray.html.md)