One-call federated query. Searches records semantically + scans KG by entity overlap with the query tokens. namespace OR namespaces: pass one for single-session recall, or a list to fan across multiple sessions (e.g., a user's last 10 sessions for the cross-session graph-RAG narrative). namespaces wins if both are provided. mode: None (default) -- pure semantic ranking by cosine score. 'qa_optimized' -- additionally BM25-fuse over the candidate pool via Reciprocal Rank Fusion (RRF). Worth ~+20pp on memory-QA benchmarks where the question contains specific entity tokens (brand names, amounts, dates) that semantic search alone under-weights. recency_first: False (default) -- semantic relevance ordering. True -- after ranking, re-sort the returned records by ts desc (newest first). Use for knowledge-update questions where the most recent fact wins.
Discovery needs no key. Ratings and market analysis are Pro.
Get an API key
Free tier, no form to fill in. Signing in shares your email address with us — we
store it to create your key and to recognise you if you sign in with another
provider. See our Privacy Policy and
Terms.