Adaptive

Intelligent LLM infrastructure that automatically selects the optimal model for each request. Drop-in OpenAI replacement with 30-70% cost savings.

Features

• 🧠 Smart routing - AI selects optimal models automatically
• 💰 Cost optimization - 30-70% savings vs direct provider usage
• ⚡ Fast caching - Dual-layer cache for instant responses
• 🔄 Multi-provider - OpenAI, Anthropic, Groq, DeepSeek, Google AI
• 📊 Analytics - Usage tracking and cost insights

Quick Start

1. Deploy

git clone https://github.com/your-org/adaptive.git
cd adaptive
cp .env.example .env  # Add your API keys
docker-compose up -d

2. Use with any OpenAI SDK

from openai import OpenAI

client = OpenAI(
    api_key="your-adaptive-key",
    base_url="https://your-deployment-url/v1"
)

response = client.chat.completions.create(
    model="",  # Empty string enables intelligent routing
    messages=[{"role": "user", "content": "Hello world"}]
)

Architecture

• adaptive-backend/ - Go API server (Fiber, OpenAI SDK, Redis)
• adaptive_ai/ - Python ML service (LitServe, HuggingFace, scikit-learn)
• adaptive-app/ - Next.js web app (React 19, Prisma, tRPC, Clerk)
• adaptive-docs/ - Documentation (Mintlify)

Development

# Backend
cd adaptive-backend && go run cmd/api/main.go

# AI Service  
cd adaptive_ai && uv run adaptive-ai

# Frontend
cd adaptive-app && bun dev

Documentation

See adaptive-docs/ for complete documentation and API reference.