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 | 0.44× | 80 / 290 | 76 / 284 | 2 / 6 |
| 2026-08-22 | 0.34× | 113 / 424 | 105 / 417 | 2 / 6 |
| 2026-08-23 | 0.33× | 108 / 295 | 104 / 292 | 2 / 6 |
| 2026-08-24 | 0.30× | 95 / 252 | 94 / 248 | 2 / 4 |
| DEPLOYMENT | $/1000 COMPLETED | PRICE $/HR | SAMPLE | PRICE PROVENANCE |
|---|---|---|---|---|
| GeForce RTX 5090 | $1.15 | $0.65 | 252 attempts | afhubbard/gpu-prices (CC BY 4.0)… |
| RTX PRO 4500 Blackwell | $2.44 | $0.72 | 70 attempts | afhubbard/gpu-prices (CC BY 4.0)… |
The measured pair page for these deployments: side-by-side cells. Downloads: ratio cells, value board.