A modern chatbot interface built with Next.js that connects to your local Ollama models via FastAPI.

```bash

brew install ollama

curl -fsSL https://ollama.ai/install.sh | sh

```

```bash

ollama serve

ollama pull llama3.2

```

```bash pip install fastapi uvicorn httpx ```

```bash

python scripts/fastapi_server.py ```

```bash npm run dev ```

Create a .env.local file: ``` FASTAPI_URL=http://localhost:8000 OLLAMA_MODEL=llama3.2 OLLAMA_BASE_URL=http://localhost:11434 ```

• 🤖 Local AI inference with Ollama
• 💾 Auto-save conversations to localStorage
• 📁 Export chats as JSON files
• 🔄 Switch between conversation sessions
• 🎨 Professional red-themed UI
• ⚡ Fast local processing (no API costs!)

1. Open this project in VSCode
2. Install the Python extension
3. Use the integrated terminal to run both servers
4. Debug FastAPI with breakpoints if needed

• Ollama not connecting: Make sure ollama serve is running
• Model not found: Run ollama pull <model-name> first
• FastAPI errors: Check the terminal running the Python server
• CORS issues: The FastAPI server is configured for localhost:3000