# Sapphire AI - Complete Technical Documentation

## Overview
Sapphire AI provides high-performance AI inference with dual operating modes (Flash and Ultra) designed for developers, engineers, and power users.

## Operating Modes
1. Flash Mode: Ultra-low latency responses optimized for real-time coding assistance and quick queries.
2. Ultra Mode: Deep context reasoning engine for complex software architecture and multi-step problem solving.

## Model Context Protocol (MCP) Integration
Sapphire AI exposes a native Model Context Protocol server for direct agent tool calling.
- Endpoint: https://sapphire-ai-main.vercel.app/.well-known/mcp
- Transport: Streamable HTTP / JSON-RPC 2.0

## Support & Contact
- Email: support@sapphire-ai.com
- Location: Karachi, Sindh, Pakistan