[General question] Rate Limits with BYOK

Hey! I was wondering how the application handles rate limits when they're hit by BYOK API providers. I'm really interested because I've been using one model very heavily and haven't hit a single rate limit, even though I usually hit dozens.

No files, tests, or entry points are identified in the issue. Start by locating the BYOK provider rate-limit handling and confirming the behavior with maintainers; done would require an agreed explanation of how provider limits are handled and where that explanation should live.

Written by the indexing model from the issue text.