store

STORE: nearest-neighbor search over vectors you upserted in...

STORE: nearest-neighbor search over vectors you upserted in a namespace, scoped to YOUR wallet. Pass a query embedding (same dimensionality as the stored vectors) and get the top-K most similar by cosine similarity, each with its body + meta + score (1.0 = identical). The retrieval half of agent memory / RAG — bring your own query embedding. Querying does NOT extend TTLs. Private to your wallet (the x402 payer).

post/api/store/vector-query

Request

  • Base URL: https://2s.io
  • URL: https://2s.io/api/store/vector-query
  • Auth: API key in header PAYMENT-SIGNATURE

Request body

nsstring required

Namespace (index) to search.

embeddingnumber[] required

Query embedding (float array), same dimensionality as the stored vectors.

topKinteger

How many neighbors to return. Default 10, max 100.

Response

(unresolved $ref)

Changes

No recorded changes to this operation across all 1 revision of this API.