Repolode
Submit
MabudAlam

MabudAlam/NomAI-App

A Flutter nutrition and meal-tracking app backed by a FastAPI and LangChain agent that analyzes food photos and builds weekly diet plans.

Open-source alternative to:MyFitnessPalLose It!

View on GitHub
Stars
126
Forks
15
Open issues
0
Last commit
6 months ago

What is it?

NomAI is an AI nutrition assistant with a Flutter client for iOS, Android and web, paired with a FastAPI backend that orchestrates LLM calls. Users can photograph a meal or describe it in text, and the app runs a three-step pipeline: extract the food items, search the web via Exa or DuckDuckGo for nutrition data, then have the LLM synthesize structured nutritional information.

Beyond analysis, the project includes a LangChain-based chat agent that takes dietary preferences, allergies and health goals into account, a weekly diet planner that generates seven-day meal plans with carb cycling and macro targets, and suggestions for five alternative meals. Meals can be marked as eaten and plans updated, with chat history and diet plans stored in Google Firestore.

The backend can use either Google Gemini or OpenRouter as the LLM provider, selected through environment variables. The client relies on Firebase Authentication and Remote Config to point at the deployed backend.

Key features

  • 01Analyzes meals from photos or text descriptions using a web-grounded LLM pipeline
  • 02Chat assistant that accounts for dietary preferences, allergies and health goals
  • 03Generates 7-day diet plans with carb cycling and macro targets
  • 04Suggests five alternative meals that fit the user's dietary profile
  • 05Switches between Google Gemini and OpenRouter as the LLM provider
  • 06Stores chat history and diet plans in Google Firestore

Use cases

  • Logging meals by snapping a photo and getting an estimated nutrition breakdown
  • Building a weekly meal plan that respects allergies and macro goals
  • Studying a reference implementation of a FastAPI and LangChain nutrition agent with a Flutter client

Topics

#flutter#fastapi#langchain#nutrition#meal-planning#llm#firebase#ai#calai#langraph

FAQ

What is NomAI-App?
A Flutter nutrition and meal-tracking app backed by a FastAPI and LangChain agent that analyzes food photos and builds weekly diet plans. NomAI is an AI nutrition assistant with a Flutter client for iOS, Android and web, paired with a FastAPI backend that orchestrates LLM calls. Users can photograph a meal or describe it in text, and the app runs a three-step pipeline: extract the food items, search the web via Exa or DuckDuckGo for nutrition data, then have the LLM synthesize structured nutritional information.
Is NomAI-App free to use? What license does it have?
NomAI-App has no declared license. Its code is public, but without a license you have no explicit permission to reuse it; ask the author before relying on it.
Is NomAI-App still maintained?
Updated within 6 months. The last commit to NomAI-App was on Apr 22, 2026.
What can I use NomAI-App for?
Logging meals by snapping a photo and getting an estimated nutrition breakdown Building a weekly meal plan that respects allergies and macro goals Studying a reference implementation of a FastAPI and LangChain nutrition agent with a Flutter client
Where can I get NomAI-App?
The source code and installation instructions are on GitHub: https://github.com/MabudAlam/NomAI-App

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects