Why Aren't Openrouter.ai Models Displayed in n8n Cloud Despite Valid Credentials?

Below is a list of things to verify and try when the models list does not load despite valid credentials:

1. Check the Amended Base URL
   The node fetches the available models from the API endpoint (typically “/v1/models”). When using an alternative endpoint (as with Openrouter.ai), confirm that the amended base URL points to an endpoint that returns the expected structure. You may want to test the endpoint manually (for example via an HTTP Request node) to validate that it returns a list of models.

2. Confirm API Compatibility
   Ensure that Openrouter.ai’s API is compatible with what n8n expects when loading models. Some platforms require additional parameters or have minor differences in endpoint responses. Comparing the response of an HTTP Request to Openrouter.ai’s endpoint against the documented OpenAI endpoint might reveal discrepancies.

3. Try Using HTTP Request Node for Testing
   Create a temporary workflow that uses an HTTP Request node to query the models endpoint. This helps confirm whether the endpoint is returning valid data. For example:

   JSONCopied
   {
     "nodes": [
       {
         "parameters": {
           "requestMethod": "GET",
           "url": "https://your-amended-base-url/v1/models"
         },
         "name": "FetchModels",
         "type": "n8n-nodes-base.httpRequest",
         "typeVersion": 1,
         "position": [450, 300]
       }
     ]
   }

   Replace "https://your-amended-base-url" with the one you are using. If the response data is not what n8n expects (or if there is an error), that would explain why no models are selectable.

4. Double-Check Credentials and Organization Settings
   Even though the credentials validate, verify that any required headers (for example, if Openrouter.ai needs a specific organization ID or custom header) are correctly configured. Documentation for OpenAI credentials specifies using an API key and optionally an Organization ID. Check whether any additional changes are required when using an alternative provider.

5. Review n8n Logs and Additional Error Details
   Although the interface shows only a red triangle warning, you may be able to get additional error details from the logs. This can sometimes reveal if the error is due to an unexpected response structure or another configuration issue.

6. Contact n8n Support or Check the Community Forum
   As this issue involves a 3rd-party amendment (and the integration originally comes from the OpenAI Chat Model node), it can help to ask in the community forum and/or contact n8n support. Be sure to share the details (n8n version, operating system, workflow configuration, modified base URL, etc.) so that others who have used the amended endpoint can offer advice.

By confirming that the amended endpoint returns the expected models list and ensuring that all required credentials and headers are set appropriately, you should be able to narrow down the cause of the red triangle warning.

Let me know if you need further guidance on any of these steps!