GPUDECK · WORKLOAD EVIDENCE
A paid production workload on the reference fleet, measured on GeForce RTX 4080 and GeForce RTX 5090 on the same days. One workload, its own cells — never averaged with any other workload, because cross-workload aggregation is the substitution error this project exists to expose.
| DAY | GOODPUT GeForce RTX 4080 / GeForce RTX 5090 | ATTEMPTS | COMPLETIONS | CARDS |
|---|---|---|---|---|
| 2026-08-21 | 0.46× | 120 / 390 | 118 / 375 | 2 / 6 |
| 2026-08-22 | 0.44× | 177 / 518 | 169 / 503 | 2 / 6 |
| 2026-08-23 | 0.43× | 131 / 417 | 123 / 402 | 2 / 6 |
| 2026-08-24 | 0.44× | 144 / 58 | 133 / 55 | 2 / 6 |
| DEPLOYMENT | $/1000 COMPLETED | PRICE $/HR | SAMPLE | PRICE PROVENANCE |
|---|---|---|---|---|
| GeForce RTX 5090 | $5.63 | $0.65 | 281 attempts | afhubbard/gpu-prices (CC BY 4.0)… |
| RTX PRO 4500 Blackwell | $10.71 | $0.72 | 111 attempts | afhubbard/gpu-prices (CC BY 4.0)… |
The measured pair page for these deployments: side-by-side cells. Downloads: ratio cells, value board.