Comparison · GPU vs GPU · read 00:32 UTCRTX 4090 vs RTX 5080
Specifications, today's prices from 35 providers, price per unit of memory and compute, and who rents each cheapest. Better values are highlighted; lower is better for prices.
Side by sideSpecs, prices, value
Highlighted cells are the better of the two: more memory and compute, lower prices, lower cost per unit.
| RTX 4090 | RTX 5080 | Difference | |
|---|---|---|---|
| Memory & compute | |||
| VRAM | 24 GB | 16 GB | +50% |
| Memory bandwidth | 1.01 TB/s | 0.96 TB/s | +5% |
| FP16 tensor | — | — | — |
| FP32 | 82.6 TFLOPS | 56.3 TFLOPS | +47% |
| Power | 450 W | 360 W | +25% |
| Prices today | |||
| Cheapest on-demand | $0.13 | $0.19 | -29% |
| Median on-demand | $0.44 | $0.29 | +54% |
| Highest on-demand | $1.20 | $0.72 | +67% |
| Cheapest spot | $0.12 | $0.13 | -11% |
| Cheapest reserved | $0.18 | $0.24 | -24% |
| Current offers (n) | 40 | 15 | +167% |
| Providers with a fresh offer | 25 | 10 | +150% |
| Value | |||
| Per GB of VRAM · hour (cheapest) | $0.006 | $0.012 | -52% |
| Per GB of VRAM · hour (median) | $0.018 | $0.018 | +2% |
| Per PFLOP FP16 · hour (median) | — | — | — |
| Per TB/s · hour (median) | $0.437 | $0.298 | +46% |
| Trend | |||
| 30-day median change | +17.2% | +15.8% | — |
| 90-day median change | — | — | — |
90 days · median on-demand price · rebased to 100How the two have moved
Both medians indexed to 100 ninety days ago, so the slopes compare even though the price levels differ.
Cheapest providers today · on-demand · current offersWho rents each one cheapest
- 1
Vast.aiTrust 781× $0.13 - 2
GPUBankTrust 931× $0.23 - 3
gcube1× $0.25
- 1Vast.aiTrust 781× $0.19
- 2
Clore.aiTrust 491× $0.20 - 3
Theta EdgeCloud1× $0.25
Verdict · computed from the numbers aboveChoose one, choose the other
Choose the RTX 4090 if…
- your model needs more than 16 GB per GPU: 24 GB against 16 GB.
- you want choice: 25 providers have a fresh offer today against 10.
Choose the RTX 5080 if…
- the hourly price matters most: median $0.29 against $0.44, 35% less.
- you serve LLMs: memory bandwidth per dollar is better ($0.298 per TB/s-hour against $0.437).
Questions about RTX 4090 vs RTX 5080
Which is cheaper to rent, the RTX 4090 or the RTX 5080?
Today the cheapest current on-demand RTX 4090 is $0.13 per GPU-hour and the cheapest RTX 5080 is $0.19; medians are $0.44 and $0.29. Cheapest excludes offers marked unavailable and counts one floor per marketplace.
How much memory do the RTX 4090 and RTX 5080 have?
24 GB of GDDR6X on the RTX 4090, 16 GB of GDDR7 on the RTX 5080, at 1.01 and 0.96 TB/s.
Where do these numbers come from?
From 55 current offers read on the providers' own pages and APIs over the last 45 days (reference 00:32 UTC), or in our October 2026 market study for the few providers our collector does not read. Specifications come from manufacturer datasheets. Every price on the GPU pages links to its receipt and shows its read date.