Use Open WebUI with Models Hub: Admin Panel Connections Setup

Connect self-hosted Open WebUI to Models Hub as an OpenAI-compatible connection under Settings, Admin, Connections, with URL and API key fields, verification, and error troubleshooting.

Open WebUI is a self-hosted, open-source web chat interface, commonly used as a personal or team-wide front end for AI models. Its admin panel supports adding multiple OpenAI-compatible connections; point one at Models Hub and every conversation in that instance draws on your Models Hub balance. This page covers the connection only, not how to deploy Open WebUI itself.

Three things you need

A Models Hub API key, the base URL https://modelsok.com/v1, and a model name. An admin configures it once and every user on the instance can use it.

Prerequisites

Open WebUI deployed, with admin login access. If not deployed yet, see the official docs.
A Models Hub API key from the Tokens page, in a group that can access the models you want.

Recommended models

Use case	Model	Pricing	Live status
First choice for everyday chat	`claude-sonnet-4-6`	Price	Availability and latency
Complex reasoning and long-form writing	`gpt-5.5`	Price	Availability and latency
Cheap and fast for high-frequency Q&A	`gpt-5.4-mini`	Price	Availability and latency

Model names are authoritative on the pricing page.

Setup

Sign in to Open WebUI as an admin and go to Settings → Admin → Connections.
Under "Manage OpenAI API Connections", click ➕ Add Connection.
Enter the URL:
```
https://modelsok.com/v1
```
Open WebUI's own examples all include /v1; keep it here too.
Paste your Models Hub key into API Key and save the connection.
If the model list doesn't populate automatically, enter the model name (e.g. claude-sonnet-4-6) in that connection's "Model IDs (Filter)" field, save, then return to the chat page and select it.

Verify

Back on the chat page, select the model you just added and send something like "introduce yourself in one sentence". A normal reply means you're connected. Then check Usage logs in the Models Hub console; a request under that model name appears within seconds, confirming traffic went through Models Hub.

Common errors

Symptom	Cause	Fix
`401` / Unauthorized	Incomplete key, stray whitespace, or a disabled token	Re-copy from the Tokens page
Model list fails to load after saving the connection	Missing `/v1` in the URL, or the network can't reach it	Confirm the URL is `https://modelsok.com/v1`; you can also set the model manually via "Model IDs (Filter)"
Added model doesn't show up on the chat page	Model IDs (Filter) not set, or the page wasn't refreshed	Enter the model name, save, and reload the chat page
The UI becomes unresponsive after saving a bad config	Malformed URL or value caused a startup issue	Restart once with `RESET_CONFIG_ON_START=true` to recover defaults, then reconfigure
Nothing shows in usage logs	The chat isn't using a model from this connection	Check which model source is currently selected in the chat

FAQ

How is it billed? Per input and output token actually consumed in each conversation, at the price on each model's pricing page. Usage logs itemize every request.

How is this different from the vendor's API directly? Same request format; only the base URL changes, and every user on this Open WebUI instance shares one key and balance.

I set OPENAI_API_BASE_URL as an environment variable but it's not taking effect — why? In Open WebUI, connections configured in the Admin panel are stored in the database and take priority over environment variables. If both are set, the Admin panel wins; to fall back to the environment variable's default, delete the connection in Connections first.