I’ve been getting the ‘Selected model is at capacity. Please try a different model’ error for the last few hours. Each time I try a different model or start a new session, my usage limit drops by like 10%
Welcome to the forum!
Please use /feedback and open a new bug report.
/feedback - Open the feedback dialog to submit feedback and optionally include logs.
I’ve noticed similar behavior a couple of times. My understanding is that the usage meter is tied to the request itself, not just successful responses. Even if the selected model is at capacity, your request may have already gone through routing or initial processing before it fails, which can still count toward usage.
If it keeps happening for several hours, it’s probably worth reporting it through the Help Center because failed requests consistently reducing your quota doesn’t seem like the intended experience. In the meantime, waiting a bit before retrying instead of repeatedly switching models may help avoid burning through your remaining usage.