"""

🧠 Intelligent Multimedia Research Assistant (IMRA)

A smart assistant that ingests documents and returns contextual, AI-generated answers using LangChain + HuggingFace + FAISS + RAG architecture.

Features

• PDF ingestion
• LLM-based QA (Retrieval Augmented Generation)
• HuggingFace LLM backend
• Optional image generation using diffusion models

Setup

1. Install dependencies

pip install -r requirements.txt

2. Add Hugging Face token to .env

HUGGINGFACEHUB_API_TOKEN=your_token_here

3. Run the app

streamlit run ui/app.py

"""

Application URL

• The application is deployed at https://multimediaresearchassistant.streamlit.app/ and is available for all.

Output before tuning the model for resume

Output after tuning the model for resume (see fine-tuning repo at https://github.com/nayanmshah/Intelligent-Multimedia-Research-Assistant-model-finetune)