H100 vs RTX 4090
H100 starts at $1.75/GPU-hour on-demand; RTX 4090 starts at $0.35/GPU-hour on-demand. Compare the exact variants, memory and published specifications below. Peak compute figures do not establish application speed.
80.0% cheaper at RTX 4090.
updated 2026-10-06. Price alone does not establish equivalent capacity or performance.
H100
RTX 4090
Spec sheet
| H100 | RTX 4090 | |
|---|---|---|
| Spec variant | H100 SXM | RTX 4090 |
| Architecture | Hopper | Ada Lovelace |
| VRAM (spec variant) | 80 GB | 24 GB |
| FP16 TFLOPS (dense) | 989 | 165 |
| Memory bandwidth | 3350 GB/s | 1008 GB/s |
| TDP | 700 W | 450 W |
| Interconnect | NVLink 900GB/s | PCIe Gen4 |
| Launched | 2022 | 2022 |
What they cost to rent (2026-10-06)
| H100 | RTX 4090 | |
|---|---|---|
| Cheapest on-demand $/hr | $1.75 | $0.35 |
| Cheapest spot/community $/hr | $0.94 | $0.16 |
| Cheapest source | Lium | TensorDock |
| Providers offering it | 34 | 5 |
Lowest recorded on-demand rate per configuration; highlighted = cheaper. Spot/marketplace tiers shown on the individual pages.
FAQ
RTX 4090 is cheaper: the H100 rents from $1.75 per GPU-hour and the RTX 4090 from $0.35 on-demand, updated 2026-10-06.
The largest recorded H100 variant has 94 GB and the largest RTX 4090 variant has 24 GB. Other variants may have less memory. The H100 has more.
Compare more rental options: the screener · price movers · methodology.