Question 1

What is OpenGPU Network?

Accepted Answer

OpenGPU Network is a decentralized GPU computing marketplace that connects GPU providers with users who need compute power for AI inference, training, rendering, and other GPU-intensive workloads. It offers 60-80% cost savings compared to traditional cloud providers.

Question 2

How does OpenGPU Network work?

Accepted Answer

GPU providers register their machines as nodes on the network. When users submit workloads, the intelligent routing system automatically matches tasks to the most suitable providers based on hardware capabilities, latency, cost, and reputation. Payments are settled on-chain using OGPU tokens.

Question 3

Who can use OpenGPU Network?

Accepted Answer

Anyone can use the network. AI developers can run inference and training workloads, studios can render 3D content and transcode video, enterprises can offload compute-intensive tasks, and GPU owners can earn rewards by providing their idle compute capacity.

Question 4

How much can I save compared to cloud providers?

Accepted Answer

OpenGPU Network typically offers 60-80% cost savings compared to centralized cloud providers like AWS, GCP, and Azure. This is achieved through a decentralized marketplace where providers compete on pricing, eliminating the overhead of centralized infrastructure.

Question 5

What is the OGPU token?

Accepted Answer

OGPU is the native utility token of the OpenGPU Network. It is used for paying compute costs, rewarding providers who complete workloads, staking to participate in the network, and governance. OGPU operates on the OpenGPU blockchain as an ORC-20 token.

Question 6

Where can I buy OGPU tokens?

Accepted Answer

OGPU tokens are available on centralized exchanges (CEX) like MEXC and Gate.io, as well as decentralized exchanges (DEX) like Uniswap. You can also use BridgeX to transfer OGPU between different blockchain networks.

Question 7

What is staking in OpenGPU?

Accepted Answer

Staking involves locking OGPU tokens to participate in the network. Providers stake tokens to increase their reputation score and demonstrate commitment, which improves their chances of receiving workload assignments and earning higher rewards.

Question 8

How do I become a GPU provider?

Accepted Answer

To become a provider, you need a GPU machine, install the OpenGPU Provider Suite software, register your node on the network, and optionally stake OGPU tokens to boost your reputation. Once set up, your node will automatically receive and execute workloads.

Question 9

What are the hardware requirements for providers?

Accepted Answer

The network supports a wide range of GPU hardware — from consumer-grade GPUs to enterprise datacenter cards. The minimum requirements depend on the workload types you want to support. Higher VRAM and more powerful GPUs can handle larger AI models and earn higher rewards.

Question 10

How much can I earn as a provider?

Accepted Answer

Earnings depend on your GPU hardware, uptime, reputation score, and market demand. Providers are paid in OGPU tokens for each completed workload. You can track your earnings and node performance through the management dApp.

Question 11

What happens if my node goes offline during a task?

Accepted Answer

The network has built-in failover mechanisms. If a provider goes offline mid-execution, the workload is automatically rerouted to another provider. Long-running tasks use checkpoints so execution can resume from the last save point rather than starting over.

Question 12

What types of workloads are supported?

Accepted Answer

OpenGPU supports a wide range of GPU workloads including: LLM inference and chatbots, AI model training and fine-tuning, image and video generation, 3D rendering and ray tracing, video transcoding, embeddings generation, RAG pipelines, and more.

Question 13

How do I submit a workload?

Accepted Answer

You can submit workloads in two ways: (1) Native Execution — interact directly with the blockchain using OGPU tokens for full decentralization and lower fees, or (2) Relay — use a simple HTTPS API endpoint with fiat billing support, ideal for enterprises.

Question 14

How is task execution verified?

Accepted Answer

OpenGPU uses Proof of Execution — an on-chain verification mechanism that cryptographically proves a workload was completed correctly before payment is released. For critical tasks, Computation Consensus allows multiple providers to validate results independently.

Question 15

What blockchain does OpenGPU use?

Accepted Answer

OpenGPU has its own EVM-compatible blockchain using the Lachesis DAG consensus mechanism. It supports ~10,000 TPS with fast finality. Mainnet Chain ID is 1071, and it's compatible with Ethereum tooling, Solidity smart contracts, and Web3 wallets like MetaMask.

Question 16

Is OpenGPU Network secure?

Accepted Answer

Yes. Workloads run in isolated containers on provider hardware. The network uses Proof of Execution for task verification, anomaly detection for fraud prevention, and on-chain reputation tracking. Provider stakes act as economic security guarantees.

Question 17

What uptime does OpenGPU guarantee?

Accepted Answer

The decentralized network architecture provides 99.9% uptime through redundancy — if one provider fails, workloads are automatically rerouted. The intelligent routing system continuously monitors provider health and avoids unreliable nodes.

Question 18

How do I integrate OpenGPU into my application?

Accepted Answer

For the simplest integration, use the Relay HTTPS endpoint — it works like any standard API. Submit workloads via POST requests and receive results. For blockchain-native integration, use the OpenGPU SDK to interact directly with smart contracts.

Question 19

What is the difference between Relay and Native Execution?

Accepted Answer

Relay is an HTTPS API that supports fiat billing and requires no blockchain knowledge — ideal for enterprises. Native Execution interacts directly with the blockchain using OGPU tokens, offering full decentralization, lower fees, and greater transparency.

FAQ

General

Token

Providers

Workloads

Technical

Integration