Reranker
LLM-powered re-ranking — LLMReranker uses Claude Haiku as a relevance judge over the top-K keyword-retrieved candidates, producing a precision-optimized order. Opt-in and fail-safe — any API error falls back to the keyword-only order.
LLM-powered re-ranking — LLMReranker uses Claude Haiku as a relevance judge over the top-K keyword-retrieved candidates, producing a precision-optimized order. Opt-in and fail-safe — any API error falls back to the keyword-only order.