Skip to main content
Open WebUI is a self-hosted AI web client. After an administrator configures an AIHubMix OpenAI-compatible connection, users can select available models from the model picker.

​
Before you start

  • Deploy Open WebUI and sign in with an administrator account.
  • Create an API key in the AIHubMix Console. It usually starts with sk-***.
  • To restrict the models shown to users, prepare their model IDs, such as gpt-5.6-terra. Otherwise, Open WebUI can load the model list automatically.

​
Configure the AIHubMix connection

​
1. Open Connections

Click the model picker at the top of the chat page. If no model is available, click Manage Connections. You can also open Connections from the administrator settings. Open connection management from the model picker On Settings > Connections, find OpenAI API. Click the add button to create a connection, or the gear icon to edit an existing connection. Open OpenAI API Connections

​
2. Enter connection details

  • URL: https://aihubmix.com/v1. If the primary address is temporarily unavailable, use https://api.inferera.com/v1.
  • Auth: Select Bearer and enter the API key created in the AIHubMix Console.
  • Provider: Keep Default.
  • API Type: Select Chat Completions.
Enter the URL and API key For Model IDs, either leave the field empty to load models from https://aihubmix.com/v1/models, or enter a model ID such as gpt-5.6-terra and click + to expose only selected models. Click Save in the dialog, then Save again on the Connections page.

​
3. Select a model and verify

Return to the chat page and open the model picker. Search for and select an AIHubMix model under External. Select an AIHubMix model Send a short message. A successful response confirms that the connection and model configuration work. Verify a model response

​
FAQ

Why is no model shown in the picker? Confirm that you clicked Save in both the connection dialog and Connections page. If Model IDs is empty, confirm that the Open WebUI service can access https://aihubmix.com/v1/models. You can also add a model ID manually and reopen the picker. Should the URL include /chat/completions? No. Enter the API root https://aihubmix.com/v1 and select Chat Completions as the API type. What should I check when a request fails? Confirm that the API key is complete, the URL includes /v1, and the model ID is correct. If needed, use https://api.inferera.com/v1. Last updated: 2026-07-15