# Overview

Clients use the OpenGPU Network to run computational tasks on distributed GPUs.

## Why OpenGPU for Clients

OpenGPU is a **task marketplace**, not a GPU marketplace. You don't manage infrastructure. Just define your task, set time limits and rewards, and let providers compete to deliver results.

* Publish tasks, not rent GPUs
* No infrastructure management
* Providers compete for your work
* Pay only for completed tasks

## How It Works

1. **Create a Source** — Define your execution environment (Docker container + requirements)
2. **Publish Tasks** — Submit jobs with payment to your Source
3. **Get Results** — Providers compete to execute; you receive results

## Tools Available

| Tool                                                   | Best For                                 |
| ------------------------------------------------------ | ---------------------------------------- |
| [Client App](/opengpu-network/ecosystem/client-app.md) | Web-based source/task management         |
| [SDK](/opengpu-network/ecosystem/sdk.md)               | Programmatic integration                 |
| [Relay](https://docs.relaygpu.com)                     | API access without blockchain (external) |

## Quick Paths

**I want to run inference quickly:** → Use [Relay](https://docs.relaygpu.com) — API-based, no blockchain knowledge needed

**I want to publish my own model:** → [Publishing Sources](/opengpu-network/for-clients/publishing-sources.md) — Create a custom Source

**I want to automate task submission:** → [Using the SDK](/opengpu-network/for-clients/using-the-sdk.md) — Programmatic access

## Costs

* Pay per task in $OGPU
* Real-time competitive pricing
* No minimum commitments
* Payment held in escrow until completion

## AI Agents

Clients can leverage AI agents to optimize task pricing and provider selection. See [AI Agents](/opengpu-network/introduction/ai-agents.md) for details.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://opengpu-network.gitbook.io/opengpu-network/for-clients/overview.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.