Configure existing applications and tools

An application can usually use Omrouter when it supports OpenAI, OpenAI-compatible, or Custom OpenAI providers.

Configuration values

Application field	Omrouter value
Provider	OpenAI, OpenAI-compatible, or Custom OpenAI
API Key	A key created on the Keys page
Base URL, API Base, or Endpoint	https://omrouter.com/v1
Model	Model name shown on the Pricing page

The model name must match the Pricing page exactly. Do not use a name copied from a provider’s website or shorten it yourself.

If the Base URL does not work

Clients handle Base URL and Endpoint fields differently. If the default address fails, try these values in order:

https://omrouter.com
https://omrouter.com/v1
https://omrouter.com/v1/chat/completions

Change only the API address between attempts; keep the API key and model name unchanged. If the application shows the final request URL, confirm that the complete address is https://omrouter.com/v1/chat/completions.

Applications that load models automatically

Some applications call /v1/models automatically. In that case, save the API address and API key, then choose a model from the application’s model menu.

If the menu is empty, enter a model name from the Pricing page manually and test one chat request.

Common problems

Problem	What to do
API key is invalid	Copy the key again and check for leading or trailing spaces.
Model not found	Copy the model name again from the Pricing page.
Requests go to api.openai.com	Make sure the custom Base URL was saved.
Settings save, but chat fails	Try the three API addresses above in order.

If the application still cannot connect, give its name, a screenshot of the settings, and the error to an agent. Ask it to follow the agent task on the home page. Hide the API key before taking a screenshot.