Skip to main content

Table of Contents

Module spoon_ai.rag.vectorstores

Module spoon_ai.rag.vectorstores.faiss_store

FaissVectorStore Objects​

class FaissVectorStore(VectorStore)

FAISS-backed local vector store (cosine via inner product + L2 norm).

Module spoon_ai.rag.vectorstores.chroma_store

Module spoon_ai.rag.vectorstores.pinecone_store

Module spoon_ai.rag.vectorstores.qdrant_store

Module spoon_ai.rag.vectorstores.registry

get_vector_store​

def get_vector_store(backend: Optional[str] = None) -> VectorStore

Return a vector store by backend name.

Backends:

  • faiss: local/offline (mapped to in-memory cosine store)
  • pinecone: cloud Pinecone (requires PINECONE_API_KEY)
  • qdrant: local/cloud Qdrant (requires qdrant-client, default http://localhost:6333)
  • chroma: local Chroma (requires chromadb)

Module spoon_ai.rag.vectorstores.base

VectorStore Objects​

class VectorStore(ABC)

query​

@abstractmethod
def query(
        *,
        collection: str,
        query_embeddings: List[List[float]],
        top_k: int = 5,
        filter: Optional[Dict] = None) -> List[List[Tuple[str, float, Dict]]]

Return per-query list of (id, score, metadata). Higher score is better.