LangChain Lab

Repo for LangChain experiments.

langserve_demo.py demonstrates how to use LangServe to generate text and APIs for the language models.

Installation

This project uses Python 3.11.

You can install LangChain and its CLI tooling including LangServe with pip:

    pip install langchain langchain-cli

Also, access to a language model is required, e.g. the OpenAI GPT models via the API:

    pip install openai

Alternatively, you can install all the dependencies from the requirements.txt file:

    pip install -r requirements.txt

The above installs the langchain CLI. It can create and manage apps and templates, for information try:

    langchain --help

You can create a new LangChain app from a template, e.g. in the demo_app directory:

    langchain app new demo_app

You can add packages from the LangChain registry, e.g. the rag-chroma-private package:

    langchain app add rag-chroma-private