Open WebUI

Open WebUI connects to any OpenAI-compatible API. Add https://api.abliteratedllm.dev/v1 as a connection and every model we serve appears in the model selector.

Placeholders: $API_BASE is https://api.abliteratedllm.dev/v1, $API_ORIGIN is https://api.abliteratedllm.dev, and $YOUR_KEY is your API key.

Add the connection

1. Go to Settings > Admin > Connections and find Manage OpenAI API Connections.
2. Click Add Connection (the plus icon).
3. Fill in:
   • URL: $API_BASE
   • API Key: $YOUR_KEY
   • Model IDs: leave empty. Open WebUI reads the list from our /v1/models endpoint.
4. Click Verify Connection. It calls /v1/models and should report that the server connection is verified.
5. Click Save.

Pick a model, for example glm-5.3-flash-abliterated, deepseek-v4.1-flash or deepseek-v4-pro, in the model selector of a new chat.

Good to know

• No trailing slash. Enter the URL exactly as shown, ending in /v1.
• Text only. Image uploads sent to the model fail with 400 unsupported_content.
• Background requests. Features that call the model on their own, such as automatic chat titles, are separate requests and are billed like any other.
• Trial accounts allow 1 concurrent request, so a background request can collide with your chat. See Rate limits and trial.

Verified against Open WebUI docs on 2026-11-03: OpenAI-Compatible.