---
id: langchain-databricks
version: "0.1.2"
license: MIT
license_treatment: permissive
maintenance: dormant
---

# langchain-databricks — An integration package connecting Databricks and LangChain

License: permissive · Maintenance: dormant · Downloads: 181.0K/mo

## What it is and what it does

langchain-databricks was an integration package that bridged LangChain applications with Databricks services—vector search, chat models, and MLflow tracking. It depended on langchain-core for the LangChain framework, databricks-vectorsearch for semantic search, mlflow for experiment tracking, and numpy and scipy for numerical operations.

However, this package is now deprecated. The maintainers have consolidated all functionality into databricks-langchain, which is the active, forward-looking package. Existing code using langchain-databricks will continue to run, but no new features or fixes are planned here; all development has moved to the replacement package.

Use it for:

- Connecting a LangChain chatbot to Databricks vector search for semantic retrieval over enterprise data
- Integrating Databricks Foundation Models as LLM backends in LangChain applications
- Logging and tracking LangChain experiments and model runs via MLflow on Databricks
- Building RAG (retrieval-augmented generation) pipelines that query Databricks vector stores

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Connects LangChain applications to Databricks services including vector search, chat models, and MLflow integration, but is now deprecated in favor of databricks-langchain.

No. Do not install this package for new projects. It is deprecated and dormant; install databricks-langchain instead, which provides all the same functionality with active maintenance. If you have existing code using langchain-databricks, plan a migration to databricks-langchain to ensure you receive future updates and bug fixes.

## Install

pip install langchain-databricks
uv add langchain-databricks
poetry add langchain-databricks

## Installing langchain-databricks

Before you install: Low install friction with straightforward dependencies, but the package is dormant—last release was 602 days ago. Maintenance has ceased in favor of the consolidated databricks-langchain package.

License in practice: MIT license permits commercial and private use with minimal restrictions, but this is a deprecated package; new projects should use databricks-langchain instead.

Quickstart:

pip install langchain-databricks
from langchain_databricks import ChatDatabricks
chat_model = ChatDatabricks(endpoint="databricks-meta-llama-3-70b-instruct")
response = chat_model.invoke("What is MLflow?")

Requires Python 3.9 or later; Databricks workspace credentials and endpoint configuration needed at runtime.

Verify before relying:

- Whether existing code using langchain-databricks will continue to work without breaking changes as Databricks evolves
- Compatibility guarantees with future versions of langchain-core, databricks-vectorsearch, and mlflow

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: dormant
- Downloads: 181.0K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

databricks langchain integration, langchain vector search databricks, databricks chat models langchain, mlflow langchain databricks, langchain databricks connector, deprecated, databricks-integration, llm-framework

[View on SkillFed](https://skillfed.io/packages/langchain-databricks) · [View on PyPI](https://pypi.org/project/langchain-databricks/)