/ai-benchmark-platform

Primary LanguageTypeScript

AI Benchmark Platform

A comprehensive platform for tracking and comparing AI model performance across various benchmarks.

Features

  • Performance Matrix showing model scores across different benchmarks
  • Detailed benchmark information pages
  • Filtering by categories and model types
  • Search functionality
  • RESTful API for data access
  • Responsive design for all devices

Tech Stack

Frontend

  • Next.js 14
  • TypeScript
  • Tailwind CSS
  • React Hooks for state management

Backend

  • Express.js
  • TypeScript
  • Prisma ORM
  • SQLite (development) / PostgreSQL (production)

Project Structure

ai-benchmark-platform/
├── frontend/               # Next.js frontend application
│   ├── src/
│   │   ├── app/           # Next.js app directory
│   │   ├── components/    # React components
│   │   └── styles/        # Global styles
│   ├── public/            # Static assets
│   └── package.json
│
├── backend/               # Express.js backend application
│   ├── src/
│   │   ├── routes/       # API routes
│   │   └── server.ts     # Server entry point
│   ├── prisma/           # Database schema and migrations
│   └── package.json

Getting Started

Prerequisites

  • Node.js 18+
  • npm or yarn
  • Git

Installation

  1. Clone the repository: ```bash git clone https://github.com/Belzedar94/ai-benchmark-platform.git cd ai-benchmark-platform ```

  2. Install backend dependencies: ```bash cd backend npm install ```

  3. Set up the database: ```bash npx prisma migrate dev npx prisma db seed ```

  4. Install frontend dependencies: ```bash cd ../frontend npm install ```

Running the Application

  1. Start the backend server: ```bash cd backend npm run dev ```

  2. In a new terminal, start the frontend: ```bash cd frontend npm run dev ```

The application will be available at:

API Documentation

Endpoints

  • `GET /api/benchmarks`: List all benchmarks
  • `GET /api/benchmarks/:id`: Get benchmark details
  • `GET /api/models`: List all models
  • `GET /api/categories`: List all categories
  • `GET /api/search`: Search benchmarks and models

Example Response

```json { "benchmarks": [ { "id": 1, "name": "GLUE", "description": "General Language Understanding Evaluation benchmark", "category": { "id": 1, "name": "Natural Language Processing" }, "scores": [ { "score": 89.3, "model": { "name": "GPT-3" } } ] } ] } ```

Development

Code Style

  • Use TypeScript for type safety
  • Follow ESLint configuration
  • Use Prettier for code formatting

Branch Strategy

  • `master`: Production-ready code
  • `develop`: Development branch
  • Feature branches: `feature/feature-name`

Testing

Run tests: ```bash npm test ```

Deployment

Frontend (Vercel)

  1. Connect your Vercel account
  2. Configure environment variables
  3. Deploy using Vercel CLI or GitHub integration

Backend

  1. Set up PostgreSQL database
  2. Configure environment variables
  3. Deploy to your preferred hosting service

Contributing

  1. Fork the repository
  2. Create a feature branch
  3. Commit your changes
  4. Push to the branch
  5. Create a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.

Roadmap

Phase 1 (Current)

  • ✅ Core platform functionality
  • ✅ Performance matrix
  • ✅ Basic filtering and search
  • 🔄 Testing implementation
  • 🔄 Deployment setup

Phase 2

  • User authentication
  • Enhanced visualizations
  • Community features
  • Performance optimizations
  • Advanced analytics

Contact

For questions or feedback, please open an issue in the GitHub repository.