# GPUniq Docs

> Index of the GPUniq Docs documentation.

## Documentation

- [Introduction to GPUniq](https://docs.gpuniq.com/introduction.md): GPU Meta-Cloud platform — rent GPUs, deploy containers, manage persistent storage, and generate text, images, video, music and speech through a unified API and Python SDK.
- [Quickstart](https://docs.gpuniq.com/quickstart.md): Set up your account, install the SDK, and launch your first GPU instance in minutes.
- [Python SDK](https://docs.gpuniq.com/sdk.md): Install and use the GPUniq Python SDK to control GPU instances, volumes, LLM models, and more programmatically.
- [GPU Marketplace](https://docs.gpuniq.com/gpu-marketplace.md): Browse 5000+ GPU servers from multiple providers, compare specs and pricing, and rent specific machines with full control.
- [GPU Dex-Cloud](https://docs.gpuniq.com/gpu-dex-cloud.md): Deploy GPU instances by type — pick a GPU model, click deploy, and the platform finds the best available server automatically.
- [GPU Burst](https://docs.gpuniq.com/gpu-burst.md): Scale to 100 GPUs with automatic fallback — deploy multi-GPU workloads across multiple providers with a single API call.
- [Managing Instances](https://docs.gpuniq.com/managing-instances.md): Create, monitor, and control your GPU instances — start, stop, delete, view logs, check SLA, and manage SSH keys.
- [LLM & Media API](https://docs.gpuniq.com/llm.md): Chat, reasoning, images, video, music, speech and upscaling through one API key and one USD balance. Claude, GPT, Gemini, Grok, DeepSeek, Qwen, Nano Banana, Seedance, Veo and more.
- [Models & pricing](https://docs.gpuniq.com/llm/models.md): 370+ models on one balance — how each media kind is billed, how discounts apply, and how to read the live catalog instead of a table that goes stale.
- [Chat & completions](https://docs.gpuniq.com/llm/chat.md): The native GPUniq chat API — completions, server-side chat sessions with full history, shareable transcripts, the generations library, and natural-language to shell commands.
- [OpenAI & Anthropic compatibility](https://docs.gpuniq.com/llm/openai-compat.md): Drop-in replacements for api.openai.com and api.anthropic.com — Chat Completions, the Responses API, and a native Messages endpoint that Claude Code talks to without a proxy.
- [Embeddings & rerank](https://docs.gpuniq.com/llm/embeddings.md): OpenAI-compatible embeddings and Cohere-style reranking on the same key and balance — models, request shapes, billing and errors.
- [Image generation](https://docs.gpuniq.com/llm/images.md): Text-to-image and image-to-image across Nano Banana, GPT Image 2, FLUX.2, Seedream, Z-Image, Kling and Midjourney — one async job API, billed per delivered image.
- [Video generation](https://docs.gpuniq.com/llm/video.md): Text-to-video, image-to-video, first/last-frame and reference-to-video across Veo 3.1, Kling, Seedance, Sora, Wan and Runway — one async job API, charged on delivery.
- [Music & speech](https://docs.gpuniq.com/llm/audio.md): Text-to-music with lyrics and style, plus ElevenLabs text-to-speech, multi-speaker dialogue, sound effects and voice isolation — two async job surfaces, billed on delivery.
- [Topaz upscaling](https://docs.gpuniq.com/llm/topaz.md): Topaz Labs image enhancement (Gigapixel / Wonder / Denoise / Sharpen) and video upscaling (Proteus / Starlight / frame interpolation) through the GPUniq media API — credit-metered billing, job-based delivery, pay only for delivered results.
- [Long generations](https://docs.gpuniq.com/llm/long-generations.md): Streaming, job-based polling, and the keep-alive path for long non-streaming requests — how to make GPUniq generate 5000 to 100,000 tokens reliably without hitting the edge-proxy timeout.
- [Error reference](https://docs.gpuniq.com/llm/errors.md): Complete error catalog for the GPUniq LLM API — HTTP status codes, OpenAI error.type / error.code mappings, what triggers each error, and the right way to recover.
- [Volumes](https://docs.gpuniq.com/volumes.md): Persistent storage that survives instance restarts — upload data, attach volumes to any GPU instance.
- [Settings](https://docs.gpuniq.com/settings.md): Manage SSH keys for passwordless access and configure Telegram notifications.
- [CLI Client](https://docs.gpuniq.com/cli-client.md): Rent GPUs, SSH into them, chat with LLMs, generate images, and manage volumes — all from your terminal with the `gg` command.
- [CLI on GPU Instance](https://docs.gpuniq.com/cli-gpu.md): Command-line tool for checkpointing, replaying, and tracking commands on GPU instances.

## API Reference

- [Login](https://docs.gpuniq.com/post-v1-auth-login.md): Authenticate and receive JWT tokens.
- [Get current user](https://docs.gpuniq.com/get-v1-auth-me.md): Get profile of the authenticated user.
- [Google OAuth](https://docs.gpuniq.com/get-v1-auth-google.md): Initiate Google OAuth authentication flow.
- [List available GPUs](https://docs.gpuniq.com/get-v1-marketplace-list.md): Browse available GPUs with filtering options.
- [Get GPU details](https://docs.gpuniq.com/get-v1-marketplace-agent-agent_id.md): Get detailed information about a specific GPU/agent.
- [Create rental order](https://docs.gpuniq.com/post-v1-marketplace-order.md): Rent a GPU. Returns SSH credentials upon success.
- [Create rental order (async)](https://docs.gpuniq.com/post-v1-marketplace-order-async.md): Create order asynchronously. Returns job_id for status polling.
- [Get order status](https://docs.gpuniq.com/get-v1-marketplace-order-status-job_id.md): Poll for async order creation status.
- [List my instances](https://docs.gpuniq.com/get-v1-instances-my.md): Get list of your active GPU rentals.
- [Get instance details](https://docs.gpuniq.com/get-v1-instances-task_id.md): Get detailed information about a specific rental.
- [Delete instance](https://docs.gpuniq.com/delete-v1-instances-task_id.md): Terminate a rental permanently.
- [Start instance](https://docs.gpuniq.com/post-v1-instances-task_id-start.md): Resume a stopped instance.
- [Stop instance](https://docs.gpuniq.com/post-v1-instances-task_id-stop.md): Stop a running instance to pause billing.
- [Create deposit (YooKassa)](https://docs.gpuniq.com/post-v1-payments-deposit.md): Create a deposit request using YooKassa.
- [Get payment history](https://docs.gpuniq.com/get-v1-payments-history.md): Get list of all payments.
- [Get balance](https://docs.gpuniq.com/get-v1-crypto-balance.md): Get user's USDT balance.
- [Get deposit address](https://docs.gpuniq.com/post-v1-crypto-deposit-address.md): Get USDT TRC20 address for deposits.
- [Create withdrawal](https://docs.gpuniq.com/post-v1-crypto-withdrawal.md): Withdraw USDT to external wallet.
- [Get transaction history](https://docs.gpuniq.com/get-v1-crypto-transactions.md): Get crypto transaction history.
- [Get GPU recommendation](https://docs.gpuniq.com/post-v1-ai-recommendation.md): Get AI-powered GPU recommendations based on your query.
- [Health check](https://docs.gpuniq.com/get-v1-health.md): Check API health status.

## Help Center

- [Help Center](https://docs.gpuniq.com/help-center.md)
- [Frequently Asked Questions](https://docs.gpuniq.com/help-center/faq/common-questions.md): Common questions about the GPUniq GPU Meta-Cloud platform.
- [Connection Issues](https://docs.gpuniq.com/help-center/troubleshooting/connection-issues.md): How to fix connection problems with your GPUniq GPU rental.
- [How to Rent Your First GPU](https://docs.gpuniq.com/help-center/guides/rent-your-first-gpu.md): Step-by-step guide to renting a GPU on GPUniq — from account creation to running your first workload.

## Changelog

- [Changelog](https://docs.gpuniq.com/changelog.md): Track the latest updates and improvements to GPUniq.