How does yield generation work?

The deployed GPUs are operated by data center partners. They're utilized for enterprise AI workloads or sold to cloud buyers, generating revenue. That revenue is distributed as USDC yield to GNFT and miniGPU holders.

Revenue Generation Process

1. Hardware Deployment

• Professional-grade GPUs are deployed in tier-1 data centers
• Operators provide power, cooling, networking, and physical security
• Hardware is configured for optimal performance and utilization

2. Workload Execution

• AI Training: Large language models, computer vision, and machine learning training
• Inference Services: Real-time AI model serving and prediction APIs
• Cloud Computing: On-demand GPU compute for enterprises and developers
• Research Applications: Academic and scientific computing workloads

3. Revenue Collection

• Operators charge clients based on usage (per hour, per job, or subscription)
• Revenue is collected in fiat currencies and converted to USDC
• Operating expenses (power, cooling, maintenance) are deducted

Yield Distribution

Revenue Split

After operating expenses (hosting, energy, rent, operator costs, etc.), net proceeds are split:

• 90% to investors (GNFT and miniGPU holders)
• 10% to Compute Labs (management fee)

Distribution Schedule

• Accrual: Yield accrues daily based on actual GPU utilization
• Payment: Distributions are made monthly on a net 30 basis
• Currency: All yields are distributed in USDC for stability
• Gas-Free: Claiming process is optimized to minimize transaction costs

Yield Calculation

Yield is calculated based on:

• Utilization Rate: Percentage of time GPU is actively processing workloads
• Market Rates: Current pricing for GPU compute in the market
• Operating Efficiency: Cost optimization by operators and Compute Labs
• Hardware Performance: Actual computational output vs. theoretical maximums

Transparency and Monitoring

Real-Time Tracking

• Live dashboard showing GPU utilization, uptime, and performance metrics
• Revenue tracking with detailed breakdowns by workload type
• Operating expense reporting for full financial transparency

Verification Systems

• Proprietary monitoring agents on each GPU
• Third-party auditing of revenue and expense reports
• Blockchain-recorded yield distributions for immutable transparency

This systematic approach ensures that yield generation is tied to real economic activity, providing sustainable returns based on actual demand for AI compute resources.