How DePINly estimates io.net GPU earnings. Last verified 2026-05-22.
TL;DR
A Solana-based decentralized GPU cluster. Pay is demand-driven — you earn when a GPU is actually rented, plus block rewards. Through 2025 most "earnings" were inflationary block rewards ($16.9M emitted vs $5.7M real revenue in Q1-25); the Incentive Dynamic Engine (IDE, ~Q2 2026) phases block rewards out toward demand-driven, USD-pegged payouts (~50% supply cut). Suppliers also stake ~200 IO/chip (14-day cooldown).
Earning mechanics
- Base ~9 IO/GPU/hour (RTX 4090 baseline) × GPU multiplier (0.6× 3090 → 5.0× H100) × cluster bonus (1.05–1.35× for larger coordinated clusters) × demand utilization.
- Net subtracts electricity (GPU watts × $/kWh). Dominant variable: job-fill demand.
Formula (sketch)
gross = 9 × gpuMult × clusterMult × reliabilityMult × utilization × hours × IO_price
net = gross − (gpuWatts/1000 × hours × $/kWh)
Assumptions & confidence
| Assumption | Basis | Confidence |
|---|---|---|
| Base 9 IO/GPU/hr | recalibrated May 2026 to ~$1.20/hr at IO $0.135 | 🟡 |
| GPU multipliers / watts | relative perf + spec sheets | 🟡 |
| Cluster/reliability bonuses | no public source | 🔴 ESTIMATE |
| Demand utilization | user input — demand-variable | 🟡 |
| IO price | CoinGecko live | 🟢 |
NOT modeled
Exact IDE payout formula post-Q2 2026, staking opportunity cost (200 IO/chip), block-reward phase-out timing, IO volatility.
Confidence rating
🟡 — Mechanics (demand-driven, IDE, staking) are sourced; the cluster/reliability multipliers are illustrative estimates and earnings hinge on job-fill demand.
Sources (verified 2026-05-22)
io.net docs / IDE announcements · Q1-2025 revenue vs emission figures · CoinGecko (io).
Code
src/lib/calculators/io-net-rewards.ts · src/components/calculators/io-net/IoNetCalculator.tsx · src/app/calculate/io-net/page.tsx
These figures are point estimates inside a wide band and depend on local demand, hardware, uptime and token price. They may be wrong in either direction. Nothing here is financial advice — always do your own research.
Methodology updated 2026-05-22 · View source on GitHub →