Keep receiving: "Invalid Retool AI model" when using custom AI provider

💬 Workflows
,
1
  1. My goal: Keep utilizing the custom AI provider function via OpenRouter API
  2. Issue: Suddenly I have started receiving “Invalid Retool AI model: x” error message on all of my flows where I use a custom AI. Started in the start-mid last week.
  3. Steps I've taken to troubleshoot: Doublechecked my custom AI base url, headers, schema is all ok. Duplicating a workflow, as sometimes have fixed weird issues. Adding the custom AI again.
  4. Additional info: (Cloud, Screenshots)

image
image394×449 18.3 KB

image
image422×125 6.12 KB

image
image231×431 8.14 KB

3

Hi @jakebach,

We shipped a fix today for a regression with custom provider models

Please let us know if you are still seeing any errors on your end, and we’ll look into it!

1 Like
4

Thanks for the answer @Tess! Within my existing flows and all other use-cases with custom AI (OpenRouter). It was last confirmed to be working last friday. I still receive fx.:

{
"data":null,"error":{"message":"Invalid Retool AI model: openai/o3-mini"}
}

and
{

"data":null,"error":{"message":"Invalid Retool AI model: x-ai/grok-4-fast"}

}

The issue has not been fixed yet :frowning: :-1:

5

So sorry - it looks like it got rolled back after I messaged you. I will keep you posted when the fix goes out again

6

This fix is live again - please let me know if you still see any issues