Amazon Chat Completions Server

A unified, provider-agnostic chat completions API server supporting OpenAI and AWS Bedrock.

🚀 Quick Start

Installation & Setup

# Clone and install
git clone https://github.com/teabranch/amazon-chat-completions-server.git
cd amazon-chat-completions-server
uv pip install -e .

# Configure environment
amazon-chat config set

# Start server
amazon-chat serve --host 0.0.0.0 --port 8000

Basic Usage

# Test the unified endpoint
curl -X POST http://localhost:8000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer your-api-key" \
  -d '{
    "model": "gpt-4o-mini",
    "messages": [{"role": "user", "content": "Hello!"}],
    "stream": false
  }'

🔄 Unified Endpoint

Single Endpoint for Everything

The /v1/chat/completions endpoint is the only endpoint you need. It:

Auto-detects your input format (OpenAI, Bedrock Claude, Bedrock Titan)
Routes to the appropriate provider based on model ID
Converts between formats as needed
Streams responses in real-time when requested
Returns responses in your preferred format

Format Combinations

All format combinations are supported through the unified endpoint:

Input Format	Output Format	Use Case	Streaming
OpenAI	OpenAI	Standard OpenAI usage	✅
OpenAI	Bedrock Claude	OpenAI clients → Bedrock response	✅
OpenAI	Bedrock Titan	OpenAI clients → Titan response	✅
Bedrock Claude	OpenAI	Bedrock clients → OpenAI response	✅
Bedrock Claude	Bedrock Claude	Claude format preserved	✅
Bedrock Titan	OpenAI	Titan clients → OpenAI response	✅
Bedrock Titan	Bedrock Titan	Titan format preserved	✅

📚 Documentation

Core Documentation

Getting Started - Installation, setup, and first steps
API Reference - Complete API documentation
CLI Reference - Command-line interface guide
Architecture - System design and architecture

Guides

Usage Guide - Programming examples and use cases
AWS Authentication - AWS credential configuration
Core Components - Detailed component documentation
Testing - Testing strategies and coverage

Development

Development Guide - Extending and customizing the server
Packaging Guide - Building and distributing the package

🎯 Key Features

Unified Interface

Single Endpoint: /v1/chat/completions handles everything
Auto-Detection: Intelligent format detection
Model-Based Routing: Automatic provider selection
Format Conversion: Seamless translation between formats

Enterprise Ready

Authentication: Secure API key-based authentication
Streaming: Real-time response streaming
Error Handling: Comprehensive error management
Monitoring: Request/response logging and health checks

Developer Friendly

CLI Tools: Interactive chat, configuration, and server management
OpenAI Compatible: Drop-in replacement for OpenAI Chat Completions API
Extensible: Easy to add new providers and formats
Well Tested: Comprehensive test coverage

🔗 Quick Links

📄 License

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