cortexdb-langchain

LangChain integration for CortexDB long-term memory.

LLM provider note (audit BLK-2): the canonical quickstart uses from langchain_openai import ChatOpenAI for the example agent. CortexDB itself is LLM-agnostic — the LangChain example just needs some model to drive the agent. Swap with one line:

# OpenAI (default in the docs)
from langchain_openai import ChatOpenAI
llm = ChatOpenAI(model="gpt-4o")

# Anthropic
from langchain_anthropic import ChatAnthropic
llm = ChatAnthropic(model="claude-sonnet-4-6")

# Google
from langchain_google_genai import ChatGoogleGenerativeAI
llm = ChatGoogleGenerativeAI(model="gemini-1.5-pro")

# Local / Ollama (no API key)
from langchain_community.chat_models import ChatOllama
llm = ChatOllama(model="llama3.1")

Nothing in cortexdb-langchain reads OPENAI_API_KEY. The retriever and memory classes only talk to the CortexDB v1 surface.

Install

pip install cortexdb-langchain

API base URL

The SDK and integration default to https://api-v1.cortexdb.ai (audit FRI-8). Override with CORTEXDB_API_URL if you self-host.