Requests taking more than 5s for llama 8b instant model in production on developer plan?

Thank you for reporting, added these to the issue tracker as well