Skip to main content


This Template performs RAG using OpenSearch.

Environment Setup​

Set the following environment variables.

  • OPENAI_API_KEY - To access OpenAI Embeddings and Models.

And optionally set the OpenSearch ones if not using defaults:

  • OPENSEARCH_URL - URL of the hosted OpenSearch Instance
  • OPENSEARCH_USERNAME - User name for the OpenSearch instance
  • OPENSEARCH_PASSWORD - Password for the OpenSearch instance
  • OPENSEARCH_INDEX_NAME - Name of the index

To run the default OpenSeach instance in docker, you can use the command

docker run -p 9200:9200 -p 9600:9600 -e "discovery.type=single-node" --name opensearch-node -d opensearchproject/opensearch:latest

Note: To load dummy index named langchain-test with dummy documents, run python in the package


To use this package, you should first have the LangChain CLI installed:

pip install -U langchain-cli

To create a new LangChain project and install this as the only package, you can do:

langchain app new my-app --package rag-opensearch

If you want to add this to an existing project, you can just run:

langchain app add rag-opensearch

And add the following code to your file:

from rag_opensearch import chain as rag_opensearch_chain

add_routes(app, rag_opensearch_chain, path="/rag-opensearch")

(Optional) Let's now configure LangSmith. LangSmith will help us trace, monitor and debug LangChain applications. You can sign up for LangSmith here. If you don't have access, you can skip this section

export LANGCHAIN_API_KEY=<your-api-key>
export LANGCHAIN_PROJECT=<your-project> # if not specified, defaults to "default"

If you are inside this directory, then you can spin up a LangServe instance directly by:

langchain serve

This will start the FastAPI app with a server is running locally at http://localhost:8000

We can see all templates at We can access the playground at

We can access the template from code with:

from langserve.client import RemoteRunnable

runnable = RemoteRunnable("http://localhost:8000/rag-opensearch")

Help us out by providing feedback on this documentation page: