Mayura

Intelligent AI Routing Platform

One prompt, perfect AI model selection, superior results.

🌐 Live Demo

---

What is Mayura?

Mayura is an intelligent AI routing platform that solves the problem of choosing the right AI model for your specific task. Instead of juggling multiple AI services and guessing which one to use, Mayura automatically analyzes your prompt and routes it to the most suitable AI model.

The Problem We Solve

• Choice Paralysis: With dozens of AI models available, which one should you use?
• Context Switching: Constantly switching between different AI platforms is inefficient
• Suboptimal Results: Using the wrong model for your task leads to poor outcomes
• Complexity: Managing multiple API keys, rate limits, and pricing models

Our Solution

Mayura provides a single interface to access the best AI models through intelligent routing:

---

Key Features

Intelligent Routing

• Automatic prompt analysis and classification
• Context-aware model recommendation

Multi-Model Support

• Google: Gemini 2.5 Pro, Gemini 1.5 Pro
• Meta: Llama 3.3 70B, Llama 3.1 405B
• DeepSeek: DeepSeek R1, DeepSeek V3
• ...can be extended to more models

---

🏗️ Architecture

Mayura follows a microservices architecture designed for scalability and reliability:

🔧 Tech Stack

Component	Technology	Purpose
Infrastructure	Docker, AWS(Storage), GCP(AI Models, Auth), Azure(Compute)	Deployment and scaling
Frontend	Next.js 14, React 18, TypeScript	Modern web interface
Gateway	Go 1.23, Gin, Firebase	API gateway and routing
Classifier	Python 3.9+, FastAPI, Transformers	AI model classification
Payment	Go 1.23, LemonSqueezy, DynamoDB	Subscription management
Cache	Redis 7	High-performance caching
Database	DynamoDB, Firebase Firestore	Data persistence

---

🚀 Getting Started

📋 Prerequisites

• Node.js 18+ and npm/pnpm
• Go 1.23+
• Python 3.9+
• Docker and Docker Compose

🛠️ Installation

1. Clone the repository

   git clone --recurse-submodules https://github.com/mayuraai/mayura.git
   cd mayura

2. Set up environment variables

   # Frontend
   cp frontend/.env.example frontend/.env
   
   # Backend services
   cp backend/gateway/.env.example backend/gateway/.env
   cp backend/payment/.env.example backend/payment/.env

   Obtain your Firebase service account JSON file from the Firebase console. Copy this file to both backend/gateway and backend/payment directories, and rename it to firebase_env.json in each location.

3. Install dependencies for frontend

   # Frontend
   cd frontend
   npm install
   # or
   pnpm install

4. Start backend services with Docker

   cd backend
   docker-compose up -d

5. Start the frontend

   cd frontend
   npm run dev

6. Access the application

   • Frontend: http://localhost:3000
   • Gateway API: http://localhost:8080
   • Classifier API: http://localhost:8000

---

🛡️ Security

Security Features

• JWT Authentication with refresh tokens
• Rate limiting per user and tier
• CORS protection with allowlist
• Request/response logging

---

📜 License

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

Built with ❤️ by the Mayura Team

Pavan Manish • Sai Vishal

⭐ If you found this helpful, please give us a star!