OpenAI & LangChain chatbot with custom context knowledge base

To learn how to use this repo watch this tutorial video

This chatbot uses Pinecone index to retrieve relevant information from a custom knowledge base. Learn how to create a Pinecone index in this tutorial

Tech stack: LangChain, Pinecone, Typescript, Openai, Next.js, Tailwind

Getting Started

1. Clone the repo

git clone [github https url]

2. Install packages

npm install

3. Set up your .env file

• Change .env.example into .env and fill the neccessary keys.
• For OPENAI API key visit openai
• For Pinecone API key and environment and index names visit pinecone and create your index

4. Run the development server:

npm run dev

Deploy on Vercel

!! If your Vercel app throws back a timeout error when deployed, the response is taking more than 5s, which is the limit of a free Vercel plan.

The easiest way to deploy your Next.js app is to use the Vercel Platform

Check out Next.js deployment documentation for more details.