> For the complete documentation index, see [llms.txt](https://supernet.gitbook.io/supernet/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://supernet.gitbook.io/supernet/decentralized-gpu-system.md).

# Decentralized GPU System

<figure><img src="/files/nV1i8KwrPX8YZZZKGPce" alt="" width="375"><figcaption><p>Decentralized GPU</p></figcaption></figure>

## **Supernet Computing Power Credit System**

Supernet provides an innovative computing power credit system designed to efficiently integrate distributed computing resources, offering developers and users a scalable, low-cost, high-performance computing solution. By combining AI OS with decentralized networks (such as node communication protocols), Supernet has built a unique computing power credit ecosystem.

## **Core Architecture and Functions**

1. **Computing Power Provision and Integration**Supernet integrates globally distributed computing resources into a unified computing pool through its node network. Users can perform distributed computing for tasks such as AI model training, with computing power consumption for each task being recorded and converted to credits in real-time.
2. **Credit Calculation and Distribution**
   * **Computing Credit Rules**: Each user receives corresponding computing credits based on their contributed computing resources (such as GPU time, task complexity).
   * **Credit Usage**: Computing credits can be used to pay for network services, exchange for resource usage rights, or traded for economic returns.
   * **Dynamic Balance Mechanism**: Supernet uses smart contracts to dynamically adjust credit distribution weights, ensuring fair credit incentives for users with high contributions and demands.
3. **Task Distribution and Execution**Leveraging its distributed architecture, Supernet's computing task distribution system automatically assigns computational tasks to optimal nodes based on task scale, complexity, and resource availability. The computing power provided by AI OS ensures task execution efficiency and security.

## **System Advantages**

1. **Efficient Computing Resource Utilization**Through distributed networks and heterogeneous GPU-supported high-performance computing, Supernet significantly improves computing power efficiency while reducing costs.
2. **Incentivization and Autonomy**The credit system encourages more users to contribute computing power, supporting a positive cycle of resource supply and demand while achieving high network autonomy.
3. **Seamless Developer Experience**Supernet's computing credit system seamlessly integrates with existing blockchain protocols and applications, allowing developers to quickly deploy and expand their application scenarios.

## **Application Scenarios**

* **AI Model Training and Inference**: Developers can quickly complete large-scale training tasks based on distributed node computing support.
* **3D Rendering and Video Processing**: Users can complete complex rendering tasks by paying with computing credits.
* **Decentralized Computing Market**: Supernet's credit system supports computing power trading, building a bridge between computing power providers and users.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## 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, and the optional `goal` query parameter:

```
GET https://supernet.gitbook.io/supernet/decentralized-gpu-system.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.