Are you using LangGraph tool nodes? I run very small, non-RLHF instruction models with maybe a 2% failure rate on response format matching tool definition. I wo

Are you using LangGraph tool nodes? I run very small, non-RLHF instruction models with maybe a 2% failure rate on response format matching tool definition. I would also guess you do not have your orchestration and pipe configured correctly.