---
title: Persian AI Assistant
emoji: 🧠
colorFrom: blue
colorTo: indigo
sdk: docker
sdk_version: "3.0.0"
app_file: app.py
pinned: false
---

# Persian AI Assistant

A FastAPI-based AI assistant with real-time chat capabilities, designed specifically for Persian language interaction. This project uses the BLOOM model for natural language processing and provides a modern, responsive web interface.

## Features

- 🤖 AI-powered conversation in Persian
- 💬 Real-time WebSocket communication
- 🌙 Dark/Light theme support
- 📱 Responsive design
- ⚡ Fast and efficient processing

## Technical Stack

- FastAPI for backend API
- WebSocket for real-time communication
- PyTorch & Transformers for AI processing
- Docker for containerization
- Modern HTML/CSS/JS frontend

## Environment Variables

The following environment variables are configured automatically:
- `HF_HOME`: Cache directory for Hugging Face models
- `TRANSFORMERS_PARALLELISM`: Parallelism settings
- `TORCH_HOME`: PyTorch models directory

## Resources

The application is optimized for Hugging Face Spaces and uses:
- CPU-optimized PyTorch
- Efficient memory management
- Cached model loading

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference