langchain-turbopuffer

This repository contains the langchain-turbopuffer integration package for LangChain.

turbopuffer is a fast, cost-efficient vector database. This package provides a TurbopufferVectorStore class that integrates with LangChain's vector store interface.

Installation

pip install langchain-turbopuffer

Usage

from langchain_turbopuffer import TurbopufferVectorStore
from langchain_openai import OpenAIEmbeddings
from turbopuffer import Turbopuffer

tpuf = Turbopuffer(
    # Pick the right region https://turbopuffer.com/docs/regions
    region="gcp-us-central1",
    # This is the default and can be omitted
    api_key=os.environ.get("TURBOPUFFER_API_KEY"),
)

ns = tpuf.namespace("example")

vector_store = TurbopufferVectorStore(
    namespace=ns,
    embedding=OpenAIEmbeddings(),
)

Development

See libs/langchain-turbopuffer/ for the package source.