---
title: YouTwo Memory Agent
author: MrTanTan & swkasula
colorFrom: yellow
colorTo: purple
sdk: gradio
sdk_version: 5.33.1
app_file: app.py
pinned: false
short_description: A agent with memory tools. Upload and query away!
tags:
  - agent-demo-track
  - agent
  - smolagents
  - convex
  - vectara
  - nebius
license: mit
---

Authors: [MrTanTan](https://huggingface.co/MrTanTan) && [swkasula](https://huggingface.co/MrTanTan)

Video submission:https://youtu.be/HTngofSD8zE

Full-featured project repository: https://github.com/TanGentleman/YouTwo
- Our website: https://you-two-delta.vercel.app/
- This app is also deployed using modal: https://tangentleman--youtwo-gradio-gradio-app.modal.run/


This project is made with the help of:
- [Gradio](https://gradio.app)
- [Hugging Face](https://huggingface.co)
- [Modal](https://modal.com)
- [Vectara](https://vectara.com)
- [Nebius](https://nebius.com)
- [Convex](https://convex.dev)
- And open-source contributions from the community!

# Add environment variables to your .env file
```bash
NEBIUS_API_KEY="your_nebius_api_key"
VECTARA_API_KEY="your_vectara_api_key"
```

# Run this app locally
```bash
git clone https://github.com/TanGentleman/YouTwo
cd YouTwo
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
python run_gradio_app.py
```

# Deploy the server
```bash
cd YouTwo
modal deploy run_modal_app.py
```



# YouTwo Agent - Modal Deployment Guide

This repository contains the YouTwo Agent, a conversational AI agent powered by FastRTC for real-time audio communication.

## Prerequisites

1. Install the Modal CLI:
```bash
pip install modal
```

2. Set the following secrets on Modal:
```bash
modal secret create nebius NEBIUS_API_KEY=your_nebius_api_key
modal secret create vectara VECTARA_API_KEY=your_vectara_api_key
```

## Project Structure
- `run_modal_app.py`: The main Modal deployment file
- `yt_agent`: Core agent logic
- `yt_rag`: Retrieval-augmented generation module
- `yt_gradio_app`: Gradio blocks for the project

## Deployment Instructions

### Local Development

To run the app locally during development:

```bash
modal serve run_modal_app.py
```

This will provide you with a URL to access your app during development.

### Production Deployment

To deploy the app to Modal for production:

```bash
modal deploy run_modal_app.py
```

After deployment, you'll receive a URL where your application is accessible.

## Environment Variables

The application requires the following environment variables:

- `NEBIUS_API_KEY`: API key for accessing the Nebius LLM
- `VECTARA_API_KEY`: API key for Vectara RAG system

These should be set up as secrets in Modal.

## License

This project is licensed under the MIT License. Have fun!