Skip to main content
An OpenAI connection in Open WebUI’s admin settings gives your users NagaAI’s chat models in the model picker, served through the Responses API. The fields below follow the Open WebUI docs.

​
Before you start

You need a NagaAI API key and an Open WebUI instance where you’re an admin.

​
Connect Open WebUI

1

Add the connection

In the admin settings, go to Connections, add an OpenAI connection and fill in the fields.Open WebUI marks the Responses type as experimental. If a chat breaks on it, switch API Type to Chat Completions: NagaAI serves both.
2

Limit the model list

Fill in Model IDs. Without it Open WebUI shows every model in the catalog, including image, audio and embedding models.
3

Check the connection

Open a new chat, pick one of the models and send a message. The request appears in Logs in your dashboard.

​
Configure with environment variables

Environment variables set the URL and the key. Pick the Responses API type for the connection in the admin settings afterwards:
Open WebUI’s settings for image generation, speech and RAG embeddings also accept an OpenAI base URL. NagaAI serves those with the same key.

​
Troubleshooting

​
401 invalid_api_key

Open WebUI sent no key or a wrong one. Re-enter your NagaAI API key in the connection. A provisioning key doesn’t work there.

​
The model list is full of non-chat models

The connection has no Model IDs. List the chat models you want, as in the second step.