# Runcrate Documentation

> Runcrate is the complete platform for AI teams to access open-source models and GPU compute.

## Getting Started
- /getting-started/introduction: Platform overview and product comparison
- /getting-started/quickstart: Make your first API call in 60 seconds
- /getting-started/concepts: Core concepts and terminology

## Models API (Inference Engine)
- /models/overview: Introduction to the Models API
- /models/authentication: API keys and authentication
- /models/chat-completions: Chat completions endpoint (OpenAI-compatible)
- /models/image-generation: Image generation with Flux, SDXL, Ideogram
- /models/video-generation: Video generation with Kling, Veo, Sora
- /models/text-to-speech: Text-to-speech synthesis
- /models/speech-to-text: Speech transcription with Whisper
- /models/model-catalog: Browse 140+ available models
- /models/playground: Interactive model playground

## GPU Instances (Compute)
- /instances/overview: GPU instance types and pricing
- /instances/deploying: Deploy a GPU instance
- /instances/connecting: SSH and remote access
- /instances/managing: Start, stop, and manage instances
- /instances/troubleshooting: Common issues and solutions

## Storage
- /storage/overview: Persistent storage volumes
- /storage/creating: Create and attach volumes
- /storage/mounting: Mount volumes to instances
- /storage/file-explorer: Built-in file browser

## Dedicated Clusters
- /dedicated/overview: Reserved bare-metal clusters
- /dedicated/gpu-clusters: Cluster configurations
- /dedicated/available-gpus: H100, H200, B200, B300 availability
- /dedicated/how-it-works: Procurement and setup process
- /dedicated/faq: Frequently asked questions

## SDKs
- /sdks/overview: Official SDK overview
- /sdks/python: Python SDK installation and usage
- /sdks/typescript: TypeScript SDK for Node.js and edge
- /sdks/vercel-ai: Vercel AI SDK provider

## CLI
- /cli/overview: CLI overview and capabilities
- /cli/installation: Install on macOS, Linux, Windows
- /cli/instances: Manage instances from terminal
- /cli/access: SSH and file transfer
- /cli/volumes: Volume management
- /cli/reference: Complete command reference

## MCP Server
- /mcp/overview: Model Context Protocol server
- /mcp/setup: Configure MCP for Claude, Cursor
- /mcp/tools: Available MCP tools

## Billing
- /billing/overview: Billing and usage overview
- /billing/credits: Prepaid credit system
- /billing/payment-methods: Add payment methods
- /billing/auto-recharge: Automatic credit recharge
- /billing/pricing: Pricing for inference and compute

## Account & Security
- /account/authentication: Login and account security
- /account/workspaces: Workspace management
- /account/teams: Team members and roles
- /account/api-keys: API key management
- /account/ssh-keys: SSH key management
- /account/audit-logs: Activity and audit logs

## API Reference
- /api-reference: Auto-generated from OpenAPI spec at /api-reference/openapi.json
- Inference API: https://api.runcrate.ai/v1 (OpenAI-compatible)
- Infrastructure API: https://runcrate.ai/api/v1 (instances, storage, billing)

## Examples
- /examples/ai-saas-backend: Build a production AI backend
- /examples/rag-pipeline: Retrieval-augmented generation
- /examples/ai-video-pipeline: Video generation pipeline
- /examples/deploy-llama: Deploy Llama on GPU instances
- /examples/fine-tune-llm: Fine-tune models on your data
- /examples/comfyui-cloud: Run ComfyUI in the cloud
- /examples/ollama-cloud-gpu: Run Ollama on cloud GPUs
- /examples/jupyter-notebook-gpu: Jupyter notebooks with GPUs

## FAQ
- /faq/general: General questions
- /faq/billing: Billing questions
- /faq/models: Model availability questions
- /faq/security: Security and compliance