A: Hey there, so sorry for the lengthy delay on our response here.

To answer your question, it depends on a variety of factors including how complex your app is, how long your prompt is, and which LLM is being used. Based on the feedback we've received, it's entirely possible to use all of your monthly AI tokens in one go, especially on more ambitious app requests.

That said, we're using GPT-4.1 as the default model for now since it appears to produce the best results while keeping AI token usage relatively efficient. We're actively monitoring this and will adjust as needed.

I hope this answer was helpful!