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

# Self-host Unstructured with Docker

> An API that turns PDFs, Word files, and other documents into clean data for RAG pipelines.

An API that turns PDFs, Word files, and other documents into clean data for RAG pipelines.

This page describes the Levelrail template for Unstructured (AI). It deploys the services below as one Docker Compose app on your own server. Project site: [https://github.com/Unstructured-IO/unstructured-api](https://github.com/Unstructured-IO/unstructured-api).

Recommended memory: about 2048 MiB.

## Services, ports and volumes

| Service | Image | Container ports | Volumes |
| - | - | - | - |
| `unstructured` | `downloads.unstructured.io/unstructured-io/unstructured-api:latest` | 8000 | none |

## Environment variables

| Service | Variable | Value |
| - | - | - |
| `unstructured` | `UNSTRUCTURED_API_KEY` | Generated at deploy |

Passwords and keys marked as generated are created for you when the app is deployed and stored as secrets. Values are not shown here.

## Deploy Unstructured with Levelrail

In the dashboard, open **Apps**, choose **New app**, then **Browse templates**, and select Unstructured. Review the Compose body and deploy.

With the CLI:

```bash theme={"dark"}
levelrail-cli templates deploy unstructured --name my-unstructured
```

See the [service template catalog](/templates-and-registry) for how templates work, and [Getting started](/getting-started) if you have not installed Levelrail yet.

## More AI templates

* [AnythingLLM](/self-host/anythingllm)
* [ComfyUI](/self-host/comfyui)
* [Flowise](/self-host/flowise)
* [Flowise (Postgres + Qdrant)](/self-host/flowise-postgres)
* [Jupyter GPU Notebook](/self-host/jupyter-gpu)

All templates are listed in the [self-host gallery](/self-host).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.