Comparison · GPU vs GPU · read 00:32 UTCA100 80GB vs GH200
Specifications, today's prices from 74 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.
| A100 80GB | GH200 | Difference | |
|---|---|---|---|
| Memory & compute | |||
| VRAM | 80 GB | 96 GB | -17% |
| Memory bandwidth | 2.04 TB/s | 4.00 TB/s | -49% |
| FP16 tensor | 312 TFLOPS | 989 TFLOPS | -68% |
| FP32 | 19.5 TFLOPS | 67 TFLOPS | -71% |
| Power | 400 W | 1000 W | -60% |
| Prices today | |||
| Cheapest on-demand | $0.60 | $2.29 | -74% |
| Median on-demand | $1.92 | $2.75 | -30% |
| Highest on-demand | $5.29 | $6.50 | -19% |
| Cheapest spot | $0.57 | — | — |
| Cheapest reserved | $0.76 | — | — |
| Current offers (n) | 188 | 5 | +3660% |
| Providers with a fresh offer | 69 | 5 | +1280% |
| Value | |||
| Per GB of VRAM · hour (cheapest) | $0.008 | $0.024 | -69% |
| Per GB of VRAM · hour (median) | $0.024 | $0.029 | -16% |
| Per PFLOP FP16 · hour (median) | $6.139 | $2.779 | +121% |
| Per TB/s · hour (median) | $0.939 | $0.687 | +37% |
| Trend | |||
| 30-day median change | -3.5% | — | — |
| 90-day median change | -3.5% | — | — |
Cheapest providers today · on-demand · current offersWho rents each one cheapest
- 1
Vast.aiTrust 784× $0.60 - 2
packet.ai1× $0.89 - 3
GPUBankTrust 931× $0.95
- 1
LambdaTrust 801× $2.29 - 2
Sesterce CloudTrust 601× $2.52
- 3
SpheronTrust 531× $2.75
Verdict · computed from the numbers aboveChoose one, choose the other
Choose the A100 80GB if…
- the hourly price matters most: median $1.92 against $2.75, 30% less.
- you want choice: 69 providers have a fresh offer today against 5.
Choose the GH200 if…
- your model needs more than 80 GB per GPU: 96 GB against 80 GB.
- you serve LLMs: memory bandwidth per dollar is better ($0.687 per TB/s-hour against $0.939).
Questions about A100 80GB vs GH200
Which is cheaper to rent, the A100 80GB or the GH200?
Today the cheapest current on-demand A100 80GB is $0.60 per GPU-hour and the cheapest GH200 is $2.29; medians are $1.92 and $2.75. Cheapest excludes offers marked unavailable and counts one floor per marketplace.
How much memory do the A100 80GB and GH200 have?
80 GB of HBM2e on the A100 80GB, 96 GB of HBM3 on the GH200, at 2.04 and 4.00 TB/s.
Where do these numbers come from?
From 193 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.