Comparison · GPU vs GPU · read 00:32 UTCH200 vs H200 NVL
Specifications, today's prices from 67 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.
| H200 | H200 NVL | Difference | |
|---|---|---|---|
| Memory & compute | |||
| VRAM | 141 GB | 141 GB | 0% |
| Memory bandwidth | 4.80 TB/s | 4.80 TB/s | 0% |
| FP16 tensor | 989 TFLOPS | 835.5 TFLOPS | +18% |
| FP32 | 67 TFLOPS | 60 TFLOPS | +12% |
| Power | 700 W | 600 W | +17% |
| Prices today | |||
| Cheapest on-demand | $2.09 | $2.67 | -22% |
| Median on-demand | $4.80 | $4.14 | +16% |
| Highest on-demand | $13.67 | $8.88 | +54% |
| Cheapest spot | $0.93 | $1.33 | -30% |
| Cheapest reserved | $1.67 | $2.38 | -30% |
| Current offers (n) | 111 | 27 | +311% |
| Providers with a fresh offer | 55 | 12 | +358% |
| Value | |||
| Per GB of VRAM · hour (cheapest) | $0.015 | $0.019 | -22% |
| Per GB of VRAM · hour (median) | $0.034 | $0.029 | +16% |
| Per PFLOP FP16 · hour (median) | $4.853 | $4.949 | -2% |
| Per TB/s · hour (median) | $1.000 | $0.861 | +16% |
| Trend | |||
| 30-day median change | +4.8% | +27.2% | — |
| 90-day median change | +15.7% | +27.2% | — |
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
BeamTrust 761× $2.09 - 2
GPUBankTrust 931× $2.49 - 3
Airon AI8× $2.63
- 1
Vast.aiTrust 784× $2.67 - 2Airon AI1× $2.70
- 3
IONOS2× $3.26
Verdict · computed from the numbers aboveChoose one, choose the other
Choose the H200 if…
- you want choice: 55 providers have a fresh offer today against 12.
Choose the H200 NVL if…
- the hourly price matters most: median $4.14 against $4.80, 14% less.
- you serve LLMs: memory bandwidth per dollar is better ($0.861 per TB/s-hour against $1.000).
Questions about H200 vs H200 NVL
Which is cheaper to rent, the H200 or the H200 NVL?
Today the cheapest current on-demand H200 is $2.09 per GPU-hour and the cheapest H200 NVL is $2.67; medians are $4.80 and $4.14. Cheapest excludes offers marked unavailable and counts one floor per marketplace.
How much memory do the H200 and H200 NVL have?
141 GB of HBM3e on the H200, 141 GB of HBM3e on the H200 NVL, at 4.80 and 4.80 TB/s.
Where do these numbers come from?
From 138 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.