A modern, LLM-powered bill splitting application designed for ease of use in restaurants. It targets individuals and groups looking for a streamlined way to divide shared expenses from a bill, leveraging advanced AI for item recognition and distribution.

How It Works

The application utilizes Next.js 15 with the App Router for its frontend architecture. It processes uploaded bill images (PDFs) using Together.ai's Vision models and JSON mode for structured data extraction. Users then manually assign items to individuals, and the app calculates the final split.

Quick Start & Requirements

• Install dependencies: pnpm install
• Run locally: pnpm run dev
• Prerequisites:
  • Together.ai account and API key.
  • AWS account and S3 bucket configuration.
  • .env file populated with API keys.

Highlighted Details

• Leverages Together.ai for LLM capabilities, including Vision models and JSON mode.
• Employs Helicone for LLM observability and monitoring.
• Utilizes Amazon S3 for secure image storage.
• Deployed on Vercel for seamless hosting.

Maintenance & Community

No specific information on contributors, sponsorships, or community channels is provided in the README.

Licensing & Compatibility

The repository does not explicitly state a license. Users should verify licensing for commercial use or integration with closed-source projects.

Limitations & Caveats

The application requires accounts and API keys for both Together.ai and AWS S3, adding setup complexity. The bill splitting logic relies on user input for item assignment, which may be prone to manual error.