RAG using NVIDIA NIM with LangChain

Overview

This project demonstrates the power and simplicity of NVIDIA NIM (NVIDIA Inference Model), a suite of optimized cloud-native microservices, by setting up and running a Retrieval-Augmented Generation (RAG) pipeline. NVIDIA NIM is designed to streamline the deployment and time-to-market of generative AI models across various environments, including cloud platforms, data centers, and GPU-accelerated workstations. By abstracting the complexities of AI model development and leveraging industry-standard APIs, NIM makes advanced AI technologies accessible to a broader range of developers.

Getting Started

Prerequisites

• Python 3.8 or higher
• pip and virtualenv
• API key from https://build.nvidia.com/mistralai/mixtral-8x7b-instruct

Installation

1. Clone the repository

   git clone https://github.com/mickymultani/nvidia-NIM-RAG.git
   cd nvidia-NIM-RAG
   

2. Set up a virtual environment

   Create a virtual environment named nvidia:

   python -m venv nvidia
   

   Activate the virtual environment:

   • On Windows:

     nvidia\Scripts\activate

   • On macOS/Linux:

     source nvidia/bin/activate
     

3. Install dependencies

   Install the required packages using pip:

   pip install -r requirements.txt
   

4. Environment Variables

   Create a .env file in the root directory of the project, and add your NVIDIA API key:

   NVIDIA_API_KEY=your_nvidia_api_key_here
   

   Replace your_nvidia_api_key_here with your actual NVIDIA API key.

Usage

To run the project, execute the following command:

python nim.py

Contributing

Contributions to this project are welcome!

License

Distributed under the MIT License.