🐫 CaMeL Hub
δΈ­ζ–‡ Console

Open WebUI + CaMeL Hub: One API Key for GPT, Claude, Gemini & DeepSeek

Open WebUI is the most widely deployed self-hosted, open-source chat interface for large language models β€” a single Docker container gives you a private, extensible ChatGPT-style front end with conversation history, retrieval-augmented generation over your own documents, multi-user accounts, and a growing tools/plugin ecosystem. Out of the box it talks to Ollama for local models and to any OpenAI-compatible API for cloud models. Pointing that second connection at CaMeL Hub turns Open WebUI into one interface for every major provider at once: GPT-5.5, Claude Sonnet 5, Gemini 3.1 Pro, DeepSeek V4, and 480+ other models all become reachable through a single CaMeL Hub API key and a single prepaid balance β€” no separate OpenAI, Anthropic, or Google account, and no separate billing relationship with any of them. Local Ollama models and CaMeL Hub's cloud models show up in the same model picker, so self-hosted and hosted inference sit side by side in one UI.

Open WebUI β†—

  1. Create a CaMeL Hub API key

    Sign in at the CaMeL Hub console and open the API Keys (Tokens) page. Create a new key and copy it once β€” Open WebUI will store it as the credential for the connection you add in the next step.

  2. Open the Connections settings in Open WebUI

    Sign in to your Open WebUI instance, click your avatar in the top-right corner, and go to Admin Panel β†’ Settings β†’ Connections. This is where every external model provider (Ollama, OpenAI-compatible APIs) is registered for the whole instance.

  3. Add a new OpenAI API connection

    Under the "OpenAI API" section, click the + button to add a connection. Fill in the field labeled URL / Base URL with CaMeL Hub's endpoint, and the field labeled Key / API Key with the token from step 1, then toggle the connection on and save.
    URL: https://api.camel-hub.com/v1
Key: sk-your-camel-hub-key

  4. Verify the connection

    Click the small refresh/verify icon next to the connection you just added. Open WebUI sends a request to the /v1/models endpoint on that base URL β€” a successful check confirms the key and URL are both correct before you leave the settings page.

  5. Confirm the model list populated

    Open a new chat and check the model dropdown at the top. Every model CaMeL Hub exposes on that key β€” including claude-sonnet-5, gpt-5.5, gemini-3.1-pro, and deepseek-v4-flash β€” should now be listed alongside any local Ollama models you already had configured.

  6. Choose default and visible models

    Back in Admin Panel β†’ Settings β†’ Models, you can hide models you don't use, reorder the list, and set a default model that loads for every new chat β€” useful if you mostly want one model (for example gpt-5.5 or claude-sonnet-5) but still want the full catalog available on demand.

  7. Prefer environment variables? Set them at container start

    For a fully scripted or self-hosted deployment, skip the admin UI and pass the base URL and key as environment variables when you start the container β€” Open WebUI picks them up automatically as the default OpenAI API connection.
    docker run -d -p 3000:8080 \
  -e OPENAI_API_BASE_URL=https://api.camel-hub.com/v1 \
  -e OPENAI_API_KEY=sk-your-camel-hub-key \
  -v open-webui:/app/backend/data \
  --name open-webui \
  ghcr.io/open-webui/open-webui:main

Frequently asked questions

I added the connection but the model dropdown is empty. What's wrong?

This is almost always the URL or key. Confirm the Base URL is exactly https://api.camel-hub.com/v1 (with the /v1 suffix and no trailing slash), that the key was pasted without extra whitespace, and that the connection toggle is switched on. Click the connection's refresh/verify icon to re-run the check, and confirm the Open WebUI server itself has outbound HTTPS access to api.camel-hub.com if it's running on a locked-down network.

The connection shows a failed/red status when I click verify. What does that mean?

A failed check means the /v1/models request didn't get a valid response β€” usually an invalid or expired API key, a typo in the base URL, or an outbound network/firewall block from the machine running Open WebUI. It is not a CaMeL Hub outage indicator by itself; re-check the key on the console's API Keys page first.

Which model should I set as the default?

There's no single right answer β€” pick based on the chat you use most. gpt-5.5 and claude-sonnet-5 are strong general-purpose defaults for everyday chat and coding questions; deepseek-v4-flash is a fast, inexpensive option for high-volume or low-latency use; gemini-3.1-pro is a solid choice when you need very long context. You can always override the default per chat from the model dropdown.

Can I keep using local Ollama models alongside CaMeL Hub?

Yes. Open WebUI supports multiple connections simultaneously β€” your existing Ollama connection and the CaMeL Hub OpenAI-compatible connection both stay active, and models from both appear together in the same model picker. Nothing about adding CaMeL Hub disables or replaces local inference.