I have a crazy phenomenon. Although I have set GPT 4.1 in the agent, it always responds with GPT-4.1 nano model. Does anyone have any idea why this could be? https://cdn.discordapp.com/attachments/1362080173817860167/1362080174060867675/1.png?ex=68011783&is=67ffc603&hm=433782fb725c9372e5da5192a65b66425ff5377418ff4f5e8bc0883b03365c3c& https://cdn.discordapp.com/attachments/1362080173817860167/1362080174589481081/2.png?ex=68011783&is=67ffc603&hm=e8c817cc5b6ab8c9939f28ccd0e798588b4084de439b43b83cc07ff9772adf91&

I got the same problem, I even if I create a new workspace

Same thing - i think they just have it mapped wrong. I tried Mini and works fine (and actually works amazingly well…). @Shubham Bajaj

@Marketeer 4.1-mini better than 4o in your experience ?

I did a little test, and yes so far, but I’ll see soon when i listen back to a call but the latency is basically zero it’s fantastic. Nano would not follow the prompt 100% but mini seemed to, and obviously can’t test 4.1 yet

I'm gald others experience this, too. So I did nothing wrong on our side of the config 😉

This is a current bug and we are working to resolve it as soon as possible. Thank you for your patience