> ## Documentation Index
> Fetch the complete documentation index at: https://docs-apexspriteai.reliatrack.org/llms.txt
> Use this file to discover all available pages before exploring further.

# ApexSpriteAI: run private GPU-accelerated AI agents

> ApexSpriteAI lets you run, manage, and integrate local and cloud LLMs with powerful tool use via the Model Context Protocol. Get started in minutes.

Welcome to ApexSpriteAI — a sophisticated AI agent orchestration platform that puts the power of large language models directly in your hands. Whether you're running local models on dedicated GPU hardware or integrating cloud AI into your workflows, ApexSpriteAI provides the tools you need to build fast, capable, and extensible AI coding assistants.

<CardGroup cols={2}>
  <Card title="Quick Start" icon="rocket" href="/quickstart">
    Get your AI agent running in minutes with step-by-step setup instructions.
  </Card>

  <Card title="Architecture Overview" icon="diagram-project" href="/architecture-overview">
    Understand how ApexSpriteAI's components fit together.
  </Card>

  <Card title="MCP Tools" icon="wrench" href="/concepts/mcp-tools">
    Extend your AI agent with the Model Context Protocol tool system.
  </Card>

  <Card title="Model Selection" icon="microchip" href="/guides/model-selection">
    Choose the right model for your speed and capability requirements.
  </Card>
</CardGroup>

## Get up and running

<Steps>
  <Step title="Set up your local AI backend">
    Install LM Studio and load your preferred model on your GPU hardware. ApexSpriteAI supports models from 16B to 120B parameters.
  </Step>

  <Step title="Configure your network">
    Connect your local Mac or workstation to your GPU server securely over Tailscale VPN, or run everything on a single machine.
  </Step>

  <Step title="Connect Claude Code CLI">
    Point the Claude Code CLI at your local LM Studio backend to get an AI-powered coding assistant with full MCP tool support.
  </Step>

  <Step title="Add MCP tools">
    Extend your agent's capabilities by installing MCP tools for file access, web search, code execution, and more.
  </Step>
</Steps>

## Why ApexSpriteAI?

<CardGroup cols={2}>
  <Card title="Local & Private" icon="lock">
    Run AI models entirely on your own hardware. No data leaves your network.
  </Card>

  <Card title="GPU Accelerated" icon="bolt">
    Leverage NVIDIA GPU hardware for fast, low-latency inference even with large models.
  </Card>

  <Card title="Extensible via MCP" icon="plug">
    The Model Context Protocol lets your AI agent use tools like file read/write, shell commands, and custom integrations.
  </Card>

  <Card title="Open Model Ecosystem" icon="cubes">
    Works with Qwen2.5-Coder, Llama 3.3, DeepSeek, and any model compatible with LM Studio.
  </Card>
</CardGroup>