LlamaIndex integration

ChronoVec implements the LlamaIndex VectorStore contract.

Installation

pip install "chronovec[integrations]"

Basic usage

from llama_index.core import VectorStoreIndex, StorageContext
from llama_index.core.schema import TextNode
from chronovec.integrations.llamaindex import ChronoVecLlamaStore

store = ChronoVecLlamaStore(dimensions=768)
storage_context = StorageContext.from_defaults(vector_store=store)

index = VectorStoreIndex.from_documents(
    documents,
    storage_context=storage_context,
)
retriever = index.as_retriever(similarity_top_k=5)
nodes = retriever.retrieve("what is MVCC?")

Direct node operations

# Add nodes manually
nodes = [
    TextNode(text="ChronoVec uses MVCC", metadata={"topic": "architecture"}),
    TextNode(text="Bounded reclamation", metadata={"topic": "deletion"}),
]
store.add(nodes)

# Query with metadata filtering
from llama_index.core.vector_stores import MetadataFilters, ExactMatchFilter

results = store.query(
    VectorStoreQuery(
        query_embedding=embedding,
        similarity_top_k=5,
        filters=MetadataFilters(
            filters=[ExactMatchFilter(key="topic", value="architecture")]
        ),
    )
)

Snapshot reads (ChronoVec-specific)

snapshot = store.snapshot()
# ... more nodes added ...
# query as of the snapshot
results = store.query(query, snapshot=snapshot)

API surface

ChronoVecLlamaStore implements:

• add(nodes: list[BaseNode]) → list[str]
• delete(ref_doc_id: str, **kwargs) → None
• query(query: VectorStoreQuery, **kwargs) → VectorStoreQueryResult
• snapshot() → int (ChronoVec-specific)