Comparison · GPU vs GPU · read 21:35 UTCH100 vs H200
Specifications, today's prices from 130 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.
| H100 | H200 | Difference | |
|---|---|---|---|
| Memory & compute | |||
| VRAM | 80 GB | 141 GB | -43% |
| Memory bandwidth | 3.35 TB/s | 4.80 TB/s | -30% |
| FP16 tensor | 989 TFLOPS | 989 TFLOPS | 0% |
| FP32 | 67 TFLOPS | 67 TFLOPS | 0% |
| Power | 700 W | 700 W | 0% |
| Prices today | |||
| Cheapest on-demand | $1.35 | $2.09 | -35% |
| Median on-demand | $3.17 | $4.76 | -34% |
| Highest on-demand | $8.98 | $13.67 | -34% |
| Cheapest spot | $0.67 | $0.93 | -28% |
| Cheapest reserved | $1.08 | $1.67 | -35% |
| Current offers (n) | 189 | 110 | +72% |
| Providers with a fresh offer | 76 | 54 | +41% |
| Value | |||
| Per GB of VRAM · hour (cheapest) | $0.017 | $0.015 | +14% |
| Per GB of VRAM · hour (median) | $0.040 | $0.034 | +17% |
| Per PFLOP FP16 · hour (median) | $3.201 | $4.813 | -34% |
| Per TB/s · hour (median) | $0.945 | $0.992 | -5% |
| Trend | |||
| 30-day median change | +3.7% | +4.8% | — |
| 90-day median change | +9.2% | +15.7% | — |
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
GPUBankTrust 931× $1.35 - 2
GcoreTrust 801× $1.71 - 3
LiumTrust 408× $1.79
- 1
BeamTrust 761× $2.09 - 2GPUBankTrust 931× $2.49
- 3
Airon AI8× $2.63
Verdict · computed from the numbers aboveChoose one, choose the other
Choose the H100 if…
- the hourly price matters most: median $3.17 against $4.76, 34% less.
- you serve LLMs: memory bandwidth per dollar is better ($0.945 per TB/s-hour against $0.992).
- you want choice: 76 providers have a fresh offer today against 54.
Choose the H200 if…
- your model needs more than 80 GB per GPU: 141 GB against 80 GB.
Questions about H100 vs H200
Which is cheaper to rent, the H100 or the H200?
Today the cheapest current on-demand H100 is $1.35 per GPU-hour and the cheapest H200 is $2.09; medians are $3.17 and $4.76. Cheapest excludes offers marked unavailable and counts one floor per marketplace.
How much memory do the H100 and H200 have?
80 GB of HBM3 on the H100, 141 GB of HBM3 on the H200, at 3.35 and 4.80 TB/s.
Where do these numbers come from?
From 299 current offers read on providers' own pages, in our dated study or in the market report over the last 45 days (reference 21:35 UTC). Specifications come from manufacturer datasheets. Every price on the GPU pages links to its receipt and shows its read date.