RTX 4090 vs RTX 5080
RTX 4090 starts at $0.35/GPU-hour on-demand; RTX 5080 has no current on-demand rate. Compare the exact variants, memory and published specifications below. Peak compute figures do not establish application speed.
RTX 4090
RTX 5080
Spec sheet
| RTX 4090 | RTX 5080 | |
|---|---|---|
| Spec variant | RTX 4090 | RTX 5080 |
| Architecture | Ada Lovelace | Blackwell |
| VRAM (spec variant) | 24 GB | 16 GB |
| FP16 TFLOPS (dense) | 165 | 113 |
| Memory bandwidth | 1008 GB/s | 960 GB/s |
| TDP | 450 W | 360 W |
| Interconnect | PCIe Gen4 | PCIe Gen5 |
| Launched | 2022 | 2025 |
What they cost to rent (2026-10-02)
| RTX 4090 | RTX 5080 | |
|---|---|---|
| Cheapest on-demand $/hr | $0.35 | n/a |
| Cheapest spot/community $/hr | $0.16 | $0.15 |
| Cheapest source | TensorDock | n/a |
| Providers offering it | 4 | 2 |
Lowest recorded on-demand rate per configuration; highlighted = cheaper. Spot/marketplace tiers shown on the individual pages.
FAQ
Which has more VRAM, RTX 4090 or RTX 5080?
The largest recorded RTX 4090 variant has 24 GB and the largest RTX 5080 variant has 16 GB. Other variants may have less memory. The RTX 4090 has more.
Compare more rental options: the screener · price movers · methodology.