Credential problem in new version 1.74.1 (openrouter)

Questions
1

Describe the problem/error/question:

after i upgrade to the new version 1.74.1, in the OpenAI Chat Model node, the Base URL option is moved to be inside the credential editor, after configure it for (openrouter or deepseek) APIs and set the base url the available AI Models can not be found in the list.
but it is working ok if i use OpenAI credential.

even OpenRouter Credential setting is OK:

image
image905×434 14.4 KB

But no Models pulled in the list:
image

Information on your n8n setup

  • n8n version: 1.74.1
  • Database (default: SQLite): postgres
  • n8n EXECUTIONS_PROCESS setting : main
  • Running n8n via: docker compose
  • Operating system: ubuntu 22.04
2

It looks like your topic is missing some important information. Could you provide the following if applicable.

  • n8n version:
  • Database (default: SQLite):
  • n8n EXECUTIONS_PROCESS setting (default: own, main):
  • Running n8n via (Docker, npm, n8n cloud, desktop app):
  • Operating system:
3

same issue :frowning:

4

This is actually a major issue as it has now broken production workflows :frowning:

5

@samyfaisal just FYI, I have found a work around. If you go to DeepSeek V3 - API, Providers, Stats | OpenRouter (or whichever model you want to use) and copy the breadcrumb under the main model title, you can then set the model to expression in the n8n node and paste this in. So long as the base URL is set correctly in your credentials, it should work.

As you can see, it still gives a warning in the node, but it works fine. A bit of a pain, but at least working :slight_smile:

image
image1730×975 51.4 KB