JarvisClaw Rerank

Use this endpoint when you have a set of candidate documents or passages retrieved via vector search, keyword search, or any retrieval method, and want to rerank them by semantic relevance before feeding them into an LLM. Ideal for RAG pipelines where retrieval quality matters. Prefer this over embedding-based similarity when you need a dedicated cross-encoder reranker. The x402 pay-per-call model suits sporadic or agent-driven usage without subscription commitment.

• Empty document list returns an error or empty results
• Query or documents exceed token/length limits causing a 400 error
• Payment not included or insufficient USDC balance triggers a 402 Payment Required response
• Malformed request body (missing query or documents field) returns a 422 validation error
• Service unavailable or gateway timeout returns a 5xx error
• Documents list too large causes rate limit or payload error