SpellzCode — Rerank documents by relevance: reranker for RAG (MPP)
https://agent-data-tools.revmesh2074.workers.dev/ai/rerank?query=What+is+the+capital+of+France%3F&documents=%5B%22France+has+hundreds+of+cheeses.%22%2C%22Berlin+is+the+capital+of+Germany.%22%2C%22Paris+is+the+capital+and+largest+city+of+France.%22%5DOverview
Order up to 50 documents by how well each answers a query, with a relevance score from 0 to 1 (BAAI bge-reranker-base, a cross-encoder). Send documents as a JSON array or one per line; each is read up to 2,000 characters, 40,000 in all. Returns the original index, score and text, best first; top_n keeps only the best. Also available in the Cohere-style shape at /v1/rerank. Use to pick what a model reads after search or vector retrieval.