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

# Introduction

> Inference first. Compute when you need it.

export const RuncrateStyles = () => {
  if (typeof document !== 'undefined' && !document.getElementById('runcrate-overrides')) {
    const s = document.createElement('style');
    s.id = 'runcrate-overrides';
    s.textContent = `
      /* Match Runcrate's rounding scale (--radius: 0.75rem) */
      .rounded-sm { border-radius: 0.5rem !important; }   /* 8px */
      .rounded-md { border-radius: 0.625rem !important; } /* 10px */
      .rounded-lg { border-radius: 0.75rem !important; }  /* 12px */
      .rounded-l-sm { border-top-left-radius: 0.5rem !important; border-bottom-left-radius: 0.5rem !important; }
      .rounded-r-sm { border-top-right-radius: 0.5rem !important; border-bottom-right-radius: 0.5rem !important; }
      .rounded-l-md { border-top-left-radius: 0.625rem !important; border-bottom-left-radius: 0.625rem !important; }
      .rounded-r-md { border-top-right-radius: 0.625rem !important; border-bottom-right-radius: 0.625rem !important; }
      .rounded-l-lg { border-top-left-radius: 0.75rem !important; border-bottom-left-radius: 0.75rem !important; }
      .rounded-r-lg { border-top-right-radius: 0.75rem !important; border-bottom-right-radius: 0.75rem !important; }

      /* Cards: never pure white in light mode */
      .card { background-color: #fcfcfc !important; border-radius: 0.75rem !important; }
      html.dark .card { background-color: #141414 !important; }

      /* Docs hero box */
      .rc-hero { background-color: #fcfcfc; border: 1px solid #e0e0e0; }
      html.dark .rc-hero { background-color: #141414; border-color: #242424; }
      html.dark .rc-hero h1 { color: #f5f5f5; }

      /* Runcrate scrollbar — thin, transparent track, hide-until-hover thumb */
      ::-webkit-scrollbar { width: 6px; height: 6px; background-color: transparent; }
      ::-webkit-scrollbar-track { background-color: transparent; }
      ::-webkit-scrollbar-thumb { background-color: rgba(155, 155, 155, 0.5); border-radius: 10px; transition: opacity 0.3s ease; opacity: 0; }
      ::-webkit-scrollbar-thumb:hover { background-color: rgba(155, 155, 155, 0.7); }
      *:hover::-webkit-scrollbar-thumb,
      *:focus::-webkit-scrollbar-thumb,
      *:active::-webkit-scrollbar-thumb { opacity: 1; }
      * { scrollbar-width: thin; scrollbar-color: rgba(155, 155, 155, 0.5) transparent; }
    `;
    document.head.appendChild(s);
  }
  return null;
};

<RuncrateStyles />

> ## For Agents
>
> Fetch the complete documentation index at: [https://runcrate.ai/docs/llms.txt](https://runcrate.ai/docs/llms.txt)
> Use this file to discover all available pages before exploring further.

# Introducing Runcrate

Runcrate is the complete platform for AI teams to access open-source models and GPU compute. One account gives you production inference for 140+ models, on-demand GPU instances, dedicated clusters, and the SDKs to build with all of it.

<Columns cols={4}>
  <Card title="Quickstart" icon="rocket" href="/docs/getting-started/quickstart">
    Make your first API call in under 60 seconds.
  </Card>

  <Card title="Model Catalog" icon="brain" href="/docs/models/model-catalog">
    Browse 140+ open-source models across text, image, video, and audio.
  </Card>

  <Card title="SDKs" icon="code" href="/docs/sdks/overview">
    Python and TypeScript clients. Drop-in OpenAI SDK replacements.
  </Card>

  <Card title="API Reference" icon="book" href="/docs/api-reference/introduction">
    Full REST API documentation for inference and infrastructure.
  </Card>
</Columns>

## The Runcrate Platform

Everything your AI team needs: production inference, GPU compute, and dedicated clusters — all under one account and one bill.

<Columns cols={2}>
  <Card title="Inference Engine" icon="brain" href="/docs/models/overview">
    OpenAI-compatible API for 140+ open-source models. Chat, image, video, TTS, ASR — billed per token or per generation.
  </Card>

  <Card title="GPU Compute" icon="cpu" href="/docs/instances/overview">
    On-demand instances and dedicated clusters. H100, H200, B200, B300 with root SSH access.
  </Card>
</Columns>

<Columns cols={4}>
  <Card title="Models API" icon="message" href="/docs/models/chat-completions">
    Chat completions, image generation, video, TTS, and transcription endpoints.
  </Card>

  <Card title="GPU Instances" icon="server" href="/docs/instances/deploying">
    Deploy containers or VMs with dedicated NVIDIA GPUs in 60 seconds.
  </Card>

  <Card title="Storage" icon="database" href="/docs/storage/overview">
    Persistent volumes with a built-in file explorer. Data survives instance termination.
  </Card>

  <Card title="Dedicated Clusters" icon="server-2" href="/docs/dedicated/overview">
    Reserved bare-metal clusters from 16 to 128+ nodes with InfiniBand.
  </Card>
</Columns>

## Explore use cases

See how teams use Runcrate to build AI products, run inference at scale, train models, and deploy custom servers.

<Columns cols={4}>
  <Card title="AI SaaS Backend" icon="cloud" href="/docs/examples/ai-saas-backend">
    Build a production AI backend with chat, image generation, and RAG.
  </Card>

  <Card title="RAG Pipeline" icon="search" href="/docs/examples/rag-pipeline">
    Build retrieval-augmented generation with embeddings and vector search.
  </Card>

  <Card title="Fine-tune LLMs" icon="adjustments-horizontal" href="/docs/examples/fine-tune-llm">
    Fine-tune Llama, Mistral, or Qwen on your own data with GPU instances.
  </Card>

  <Card title="Video Generation" icon="video" href="/docs/examples/ai-video-pipeline">
    Generate videos with Kling, Veo, Sora, and Seedance APIs.
  </Card>
</Columns>

## Start building

<Columns cols={3}>
  <Card title="Python SDK" icon="brand-python" href="/docs/sdks/python">
    Official Python client. Drop-in replacement for the OpenAI SDK.
  </Card>

  <Card title="TypeScript SDK" icon="brand-typescript" href="/docs/sdks/typescript">
    Official TypeScript client for Node.js and edge runtimes.
  </Card>

  <Card title="Vercel AI SDK" icon="triangle" href="/docs/sdks/vercel-ai">
    First-class Runcrate provider for the Vercel AI SDK.
  </Card>
</Columns>

<Card title="MCP Server" icon="cpu" href="/docs/mcp/overview" horizontal>
  Control Runcrate from Claude, Cursor, or any MCP-compatible AI assistant. Deploy instances, manage storage, and monitor usage with natural language.
</Card>

Or use the CLI for full terminal control:

<Columns cols={2}>
  <Card title="CLI Overview" icon="terminal" href="/docs/cli/overview">
    Deploy instances, SSH in, transfer files, and manage volumes from your terminal.
  </Card>

  <Card title="CLI Installation" icon="download" href="/docs/cli/installation">
    Install on macOS, Linux, or Windows and authenticate in 30 seconds.
  </Card>
</Columns>

***

## Which product do you need?

|                | **Inference Engine**                       | **Compute**                                                        |
| -------------- | ------------------------------------------ | ------------------------------------------------------------------ |
| **Best for**   | Building AI features on open-source models | Training, fine-tuning, custom inference servers, reserved capacity |
| **Billing**    | Per token / per generation                 | Per hour (instances) · Monthly (dedicated)                         |
| **Setup time** | 60 seconds                                 | 60 seconds (instances) · 1–2 weeks (dedicated)                     |
| **Commitment** | None                                       | None (instances) · 12–24 months (dedicated)                        |
| **Access**     | Self-serve · API key                       | Self-serve (instances) · Contact sales (dedicated)                 |
| **GPUs**       | Managed for you                            | H100, H200, B200, B300, A100, L40S, RTX 4090                       |

Not sure which fits? Start with the [Inference Engine quickstart](/docs/getting-started/quickstart). Most teams never need anything else.