Fud AI is a free, open-source, privacy-first calorie and nutrition tracker for iOS and Android. It empowers users to log food intake via photos, voice, or text, leveraging a wide array of Bring-Your-Own-Key (BYOK) AI providers for analysis and personalized coaching. The primary benefit is a highly customizable, secure, and intelligent approach to health tracking without compromising user data.

How It Works

The application integrates with 13 different AI language model providers, including Gemini, OpenAI, and Claude, allowing users to select their preferred service for food recognition and AI coaching. Users can snap photos, scan barcodes, use voice input, or type food descriptions, which are then processed by the chosen AI. The system supports on-device fallbacks, including Apple Intelligence on compatible iPhones, ensuring functionality even with intermittent provider issues. Nutrition data is reviewed and edited locally before being stored on the device and optionally synced with Apple Health or Health Connect. The AI Coach provides multi-turn conversational support, offering insights based on logged data and user goals.

Quick Start & Requirements

• Installation: Clone the repository. For iOS, open ios/calorietracker.xcodeproj in Xcode and run. For Android, open android/ in Android Studio and run.
• Prerequisites:
  • iOS: Xcode, Swift 5, SwiftUI, iOS 17.6+.
  • Android: Android Studio, Kotlin, Jetpack Compose, min SDK 26 (Android 8.0).
  • AI Providers: API keys are required for most supported LLM providers (e.g., Gemini, OpenAI). Free Gemini keys are available.
• Documentation: https://fud-ai.app
• Issue Tracker: github.com/apoorvdarshan/fud-ai/issues

Highlighted Details

• Extensive AI Provider Support: Integrates with 13 LLM providers and custom OpenAI-compatible endpoints for food analysis and AI coaching.
• Privacy-Centric Design: Operates without accounts or cloud synchronization; all user data and API keys are stored locally and encrypted.
• Versatile Logging: Supports photo capture, barcode scanning, text input, voice commands, Siri Shortcuts, and dual-camera analysis.
• AI-Powered Goals & Coaching: Features AI-driven goal recalculation based on weight trends and an AI Coach for personalized dietary advice and forecasts.
• Cross-Platform Health Sync: Bidirectional data synchronization with Apple Health (iOS) and Health Connect (Android).

Maintenance & Community

The project is actively maintained, with contributions welcomed via GitHub for bug reports and feature requests. Specific community channels like Discord or Slack are not listed.

Licensing & Compatibility

• License: MIT License.
• Compatibility: The permissive MIT license allows for commercial use and integration into closed-source applications.

Limitations & Caveats

Some advanced features, such as "Adaptive Goals," are marked as experimental. Certain AI functionalities and on-device fallbacks are platform-specific (e.g., Apple Intelligence on iOS). Users must manage API keys for their chosen AI providers, which may incur separate costs.