---
title: LLM Comparison Tool
emoji: ⚖️
colorFrom: green
colorTo: blue
sdk: docker
app_port: 7860
pinned: false
---

# LLM Comparison Tool

Side-by-side comparison of Neon AI self-hosted LLM responses against commercial models (Fireworks.ai, OpenAI, Gemini). Each Neon model's persona prompt is automatically injected into comparison model queries for fair evaluation.

## Quick Start (Development)

### 1. Configure environment

```bash
cp .env.example .env
# Edit .env with your API keys (HANA guest/password works out of the box)
```

### 2. Start the backend

```bash
cd backend
pip install -r requirements.txt
uvicorn app.main:app --reload --port 8000
```

### 3. Start the frontend

```bash
cd frontend
npm install
npm start
```

The app runs at http://localhost:3000 with the API at http://localhost:8000.

## Quick Start (Docker)

```bash
docker compose up --build
```

The app runs at http://localhost:8000.

## Features

- **Neon model discovery** via HANA API with persona prompts
- **Parallel comparison** across multiple LLM providers
- **Single query** with side-by-side response blocks
- **CSV batch** upload/download for bulk comparison
- **Session history** downloadable as CSV
- **Carousel view** for multi-Neon model comparisons
- **Dark/light theme** matching Neon project styling