README.md

---
title: Open WebUI
emoji: 🤖
colorFrom: blue
colorTo: purple
sdk: docker
pinned: false
license: mit
app_port: 8080
---

# Open WebUI on Hugging Face Spaces

This Space is configured to keep secrets out of the image, restore data from a Hugging Face dataset backup, and auto-register OpenAI-compatible providers at startup.

## What you need to set

Put sensitive values in `Secrets`.
Put non-sensitive values in `Variables`.

If you want the simplest setup, you can also put everything in `Secrets`.

## Secrets

Required:

```text
WEBUI_ADMIN_EMAIL
WEBUI_ADMIN_PASSWORD
WEBUI_ADMIN_NAME
WEBUI_SECRET_KEY
```

Provider secrets:

```text
GROQ_API_KEY
CEREBRAS_API_KEY
POLLINATIONS_API_KEY
HF_INFERENCE_TOKEN
CLOUDFLARE_API_TOKEN
CLOUDFLARE_ACCOUNT_ID
GITHUB_TOKEN
OPENAI_API_KEY
ANTHROPIC_API_KEY
GOOGLE_API_KEY
MISTRAL_API_KEY
COHERE_API_KEY
DEEPSEEK_API_KEY
TOGETHER_API_KEY
NVIDIA_API_KEY
SAMBANOVA_API_KEY
OPENROUTER_API_KEY
FIREWORKS_API_KEY
XAI_API_KEY
TAVILY_API_KEY
OMDB_API_KEY
```

Backup secrets:

```text
HF_BACKUP_REPO
HF_TOKEN
```

## Variables

Recommended:

```text
HF_BACKUP_INTERVAL_SECONDS=60
```

Optional:

```text
OPENAI_API_BASE_URL=https://api.openai.com/v1
DEFAULT_PINNED_MODELS=pollinations/kimi
TASK_MODEL_EXTERNAL=pollinations/kimi
```

## Provider endpoints used by startup script

```text
Groq              https://api.groq.com/openai/v1
Cerebras          https://api.cerebras.ai/v1
Pollinations      https://gen.pollinations.ai/v1
Hugging Face      https://router.huggingface.co/v1
GitHub Models     https://models.github.ai/inference
Cloudflare AI     https://api.cloudflare.com/client/v4/accounts/<ACCOUNT_ID>/ai/v1
OpenAI            https://api.openai.com/v1
Anthropic         https://api.anthropic.com/v1
Google Gemini     https://generativelanguage.googleapis.com/v1beta/openai
Mistral           https://api.mistral.ai/v1
Cohere            https://api.cohere.ai/compatibility/v1
DeepSeek          https://api.deepseek.com/v1
Together AI       https://api.together.xyz/v1
NVIDIA            https://integrate.api.nvidia.com/v1
SambaNova         https://api.sambanova.ai/v1
OpenRouter        https://openrouter.ai/api/v1
Fireworks         https://api.fireworks.ai/inference/v1
xAI               https://api.x.ai/v1
Tavily            https://api.tavily.com
```

## Pollinations

Pollinations works without a key for anonymous calls.

If `POLLINATIONS_API_KEY` is empty, the Space still configures:

```text
https://gen.pollinations.ai/v1
```

Recommended model for coding:

```text
pollinations/kimi
```

## Backup

Backup uses the Hugging Face dataset repo named in `HF_BACKUP_REPO`.

Example:

```text
HF_BACKUP_REPO=Nerdur/webui-backup
```

The backup stores the main Open WebUI database as:

```text
webui.db
```

Default backup interval:

```text
HF_BACKUP_INTERVAL_SECONDS=300
```

Suggested value for frequent backups:

```text
HF_BACKUP_INTERVAL_SECONDS=60
```

## Setup order

1. Add all required secrets.
2. Add `HF_BACKUP_REPO` and `HF_TOKEN` if you want restore/backup.
3. Add `HF_BACKUP_INTERVAL_SECONDS` in Variables if you want frequent backups.
4. Add `DEFAULT_PINNED_MODELS=pollinations/kimi` if you want Pollinations as the default coding model.
5. Redeploy the Space.

## Notes

- The container disk is ephemeral.
- The repo auto-restores the backup on startup if backup secrets are set.
- The repo auto-registers provider endpoints on startup if the relevant secrets are present.
- If a provider secret is missing, that provider is skipped.