Multimodal RAG System for LLMs

This project implements a multimodal RAG (Retrieval-Augmented Generation) system for large language models, enabling you to query files through API endpoints. Please note that the API expects the Ollama server to be running.

Technologies Used

• FastAPI: to build the API
• LangChain: to manage language model interactions
• Chroma: to store and handle multimodal data

Installation

Clone the repo

git clone https://github.com/aabboudi/multimodal-rag-api.git
cd multimodal-rag-api

Create and activate a virtual environment

python -m venv venv
\venv\Scripts\activate

Install dependencies

pip install -r requirements.txt

Run the server

fastapi dev app/main.py

Usage

After running the server, you can send queries to the API endpoints to interact with the multimodal RAG system. Queries can be sent through the docs, Postman, or the command line.