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 | 0.34× | 65 / 104 | 50 / 102 | 2 / 6 |
| 2026-08-22 | 0.28× | 38 / 86 | 28 / 86 | 2 / 6 |
| 2026-08-23 | 0.53× | 34 / 54 | 31 / 54 | 2 / 6 |
| 2026-08-24 | 0.41× | 63 / 70 | 54 / 67 | 2 / 4 |
| DEPLOYMENT | $/1000 COMPLETED | PRICE $/HR | SAMPLE | PRICE PROVENANCE |
|---|---|---|---|---|
| GeForce RTX 5090 | $3.04 | $0.65 | 70 attempts | afhubbard/gpu-prices (CC BY 4.0)… |
| RTX PRO 4500 Blackwell | $7.22 | $0.72 | 22 attempts | afhubbard/gpu-prices (CC BY 4.0)… |
The measured pair page for these deployments: side-by-side cells. Downloads: ratio cells, value board.