FlutterOllamaChat is a proof of concept (POC) project designed to explore AI-powered chatbot interactions using Ollama as the AI backend. The system consists of a TypeScript-based REST API with MongoDB, and a Flutter front-end for user interactions.
This project aims to develop an AI model that can be hosted locally, reducing reliance on cloud-based services like ChatGPT or Gemini, thereby eliminating API costs. A key focus is on function tooling, allowing the AI to retrieve real-time data—such as the availability of doctors in a fictional hospital—and provide informed recommendations to users.
⚠️ Note: This project is based on older templates and existing codebases, so some unused code may be present. It is intended for experimental purposes only.
- AI-powered medical assistant: A chatbot that helps users find doctors based on availability.
- Function tooling integration: AI can fetch real-time doctor availability using a MongoDB database.
- Context-aware chat: Conversations are stored, and previous messages are considered for better responses.
- Local AI model: Uses Ollama instead of cloud-based LLMs to reduce costs.
- REST API: Built with TypeScript, Express, and MongoDB.
- Flutter front-end: For user-friendly interactions.
- Node.js
- Express.js
- TypeScript
- MongoDB
- Ollama (local AI model)
- Flutter
POST http://localhost:3031/api/chat/textChat
{ "device_id": "test-mobile", "message": "Yes please!", "from_chat": "mobile", "chat_id": "67a8c97ec7a9b707be0bcb59" }
- If
chat_idis provided → The entire conversation history is sent to the AI for context. - If
chat_idis missing → A new conversation is started. It will return chat_id.
- The AI model MediHelper acts as a virtual assistant at Medicare Hospital.
- It retrieves doctor availability via function tooling and provides recommendations.
- It does not prescribe medication but can suggest home remedies for minor issues.
- It may ask for user details (name, phone number) for follow-up with a doctor.
git clone https://github.com/iamthejahid/FlutterOllamaChat cd FlutterOllamaChat
yarn install
Copy the .env.example file and rename it to .env, then update values like database info.
cp .env.example .env
Make sure MongoDB is running locally or provide a connection string in .env.
yarn run dev
This project is built upon an Express, MongoDB, TypeScript REST API starter from morshedmasud. Kudos to the original author! 👏
This project is a POC (Proof of Concept) and should not be used for real medical consultations. The AI does not replace professional medical advice. Always consult a licensed healthcare professional for medical concerns.