Today I have hit my monthly PRO limits and did some tests in AUTO mode. I added this line !!! Before completing each request, write in the chat what model you are. Only the name and version, if necessary !!! to my cursor rules. In some windows it worked, in some others it didn’t.
But then I decided to turned on the billing features to test OPUS 4.1.
So reloaded the app and switched to the MAX mode. I made a very clear, detailed prompt on some non-sophisticated topic (ECE vs ROC-AUC metrics).
The quality of the response was extremely poor, like something that you’d expect from a free model. But at the end of the response, the agent revealed that it’s Sonnet 3.5:
And I paid for this 0.19$, I can see it in my dashboard.
@Mwftw thank you for the feedback and your concern about model usage.
When you select a specific model, your request gets passed to AI provider using that model only. We have checked this several times and there is no routing issue to wrong models.
So when Claude 4.1 Opus is selected you will receive the response of Claude 4.1 Opus model.
Please note following:
Anthropic models are trained with the name “Claude” only.
Opus 4.1 models were trained on data that contained info of Claude 3.5 Sonnet.
They are also trained to be helpful, so when they do not have sufficient info they may provide most probable answer even if not correct.
I’ve tested this several times, others have checked it as well with Anthropic Console and you can test it with their API. In about 30% of cases the AI answers wrong.
