---

title: ChartGPT emoji: 🌖 colorFrom: red colorTo: red sdk: docker pinned: false

<div align="center">

📊 ChartGPT

Transform Data into Insights with the Power of AI

Try it Live • Documentation • Report Bug

</div>

---

🎯 What is ChartGPT?

Forget complex charting libraries and endless documentation. ChartGPT lets you create stunning, interactive visualizations by simply asking for what you want in plain English. Upload your CSV, ask a question, and watch as AI generates the perfect chart for your data.

You: "Show me a correlation heatmap of all numeric columns"
ChartGPT: *generates beautiful visualization + Python code*

No coding required. No chart selection headaches. Just data insights at the speed of thought.

---

✨ Features

<table> <tr> <td width="50%">

🤖 AI-Powered Magic

Natural language to beautiful charts. Powered by Qwen2.5-Coder-32B-Instruct for intelligent code generation.

🎨 Dark Mode

Easy on the eyes. Toggle between light and dark themes with a single click.

📤 Drag & Drop Upload

No complicated file pickers. Just drag your CSV and you're ready to go.

🔍 Code Transparency

See exactly how your chart was created. Every visualization comes with the Python code.

</td> <td width="50%">

⚡ Real-Time Preview

Interactive data tables with sorting, filtering, and search before you even create charts.

🔄 Iterative Refinement

Not happy with the result? Ask again and refine your visualization instantly.

🐳 Deploy Anywhere

Docker-ready. Deploy to Hugging Face Spaces, Heroku, Railway, or your own server in minutes.

🎯 Smart & Fast

Built with modern tech stack for lightning-fast responses and smooth interactions.

</td> </tr> </table>

---

🚀 How It Works

<div align="center">

graph LR
    A[📄 Upload CSV] --> B[💭 Ask Question]
    B --> C[🤖 AI Generates Code]
    C --> D[📊 Beautiful Chart]
    D --> E[🔄 Refine or Export]

</div>

1️⃣ Upload Your Data

Drag and drop any CSV file (up to 5MB). Instantly preview your data in an interactive table.

2️⃣ Ask Your Question

Type naturally:

• "Show me monthly revenue trends as a line chart"
• "Create a bar chart comparing product categories"
• "Plot the distribution of customer ages"
• "Make a scatter plot of price vs. quantity sold"

3️⃣ Get Instant Results

ChartGPT analyzes your data, writes the visualization code, and renders your chart in seconds. View the generated Python code anytime.

---

🛠️ Tech Stack

Built with cutting-edge technologies for the best developer and user experience:

• 🎯 Dash — Reactive web application framework
• 🤖 ChartGPT Library — AI-powered chart generation engine
• 🎨 Dash Mantine Components — Beautiful, modern UI components
• 📊 Plotly — Interactive, publication-quality visualizations
• 📈 Dash AG Grid — Professional data tables
• 🧠 Hugging Face Qwen2.5-Coder — State-of-the-art code generation LLM

---

💻 Installation

🐍 Local Setup

Clone and run in under 2 minutes:

# Clone the repository
git clone https://github.com/yourusername/chartgpt.git
cd chartgpt

# Create virtual environment
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Set up your Hugging Face token
echo "HUGGINGFACE_API_KEY=your_token_here" > .env

# Launch the app
python app.py

🎉 Open your browser to http://localhost:8050

🐳 Docker Setup

Even easier with Docker:

# Build the image
docker build -t chartgpt .

# Run the container
docker run -p 7860:7860 chartgpt

🎉 Access at http://localhost:7860

---

🎮 Usage Tips

| Tip | Description | |-----|-------------| | 💡 Be Specific | "Line chart of sales over time" works better than "show sales" | | 🔄 Iterate | Use "Ask again" to refine your visualizations | | 📏 File Size | Keep CSVs under 5MB for optimal performance | | 🎨 Explore | Try different chart types: scatter, bar, line, pie, heatmap, etc. | | 👀 Inspect Code | Learn from the generated Python code |

---

📁 Project Structure

chartgpt/
├── 🐍 app.py                   # Main Dash application
├── 📦 requirements.txt         # Python dependencies
├── 🐳 Dockerfile              # Container configuration
├── 📂 assets/
│   ├── 🌙 logo_dark.svg       # Logo for light theme
│   ├── ☀️  logo_light.svg      # Logo for dark theme
│   └── 🎨 custom.css          # Custom styling
└── 📖 README.md               # You are here!

---

🔧 Configuration

Customize the AI model in app.py (line 464):

# Switch to different models
chart = cg.Chart(df, model="huggingface/Qwen/Qwen2.5-Coder-32B-Instruct")

# Try other options:
# model="openai/gpt-4"
# model="anthropic/claude-3-sonnet"

---

🚀 Deployment

Hugging Face Spaces (Recommended)

Perfect for sharing with the world:

1. Create a new Space on Hugging Face
2. Select Docker as the SDK
3. Push your code to the Space repository
4. Add your HUGGINGFACE_API_KEY in Settings → Repository secrets
5. Your app goes live automatically!

Other Platforms

Deploy anywhere containers run:

| Platform | Method | |----------|--------| | 🚂 Railway | Connect GitHub repo, auto-deploy on push | | 🟣 Heroku | Use included Dockerfile with Heroku Container Registry | | ☁️ AWS/GCP/Azure | Deploy as containerized web service | | 🖥️ VPS | Run with gunicorn -b 0.0.0.0:7860 app:server |

---

🤝 Contributing

We love contributions! Here's how to get started:

# Fork the repo, then clone your fork
git clone https://github.com/YOUR_USERNAME/chartgpt.git

# Create a feature branch
git checkout -b feature/amazing-feature

# Make your changes and commit
git commit -m '✨ Add amazing feature'

# Push to your fork
git push origin feature/amazing-feature

# Open a Pull Request

---

📜 License

This project is licensed under the MIT License — feel free to use it however you'd like!

---

🙏 Acknowledgments

Standing on the shoulders of giants:

• 🎯 ChartGPT Library — Core AI chart generation
• ⚡ Plotly Dash Team — Amazing web framework
• 🎨 Mantine Dev — Beautiful component library
• 🧠 Qwen Team — Powerful open-source LLM

---

🔗 Connect

<div align="center">

</div>

---

<div align="center">

Made with ❤️ and AI

Transform your data, one question at a time

</div>