Comparison · GPU vs GPU · read 00:18 UTCH100 NVL vs Gaudi 2
Specifications, today's prices from 17 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 NVL | Gaudi 2 | Difference | |
|---|---|---|---|
| Memory & compute | |||
| VRAM | 94 GB | 96 GB | -2% |
| Memory bandwidth | 3.94 TB/s | 2.45 TB/s | +61% |
| FP16 tensor | 835.5 TFLOPS | — | — |
| FP32 | 60 TFLOPS | — | — |
| Power | 400 W | 600 W | -33% |
| Prices today | |||
| Cheapest on-demand | $1.88 | $1.04 | +80% |
| Median on-demand | $3.94 | $1.18 | +233% |
| Highest on-demand | $9.00 | $1.32 | +582% |
| Cheapest spot | $0.53 | — | — |
| Cheapest reserved | $1.92 | $0.51 | +274% |
| Current offers (n) | 29 | 7 | +314% |
| Providers with a fresh offer | 13 | 4 | +225% |
| Value | |||
| Per GB of VRAM · hour (cheapest) | $0.020 | $0.011 | +84% |
| Per GB of VRAM · hour (median) | $0.042 | $0.012 | +240% |
| Per PFLOP FP16 · hour (median) | $4.717 | — | — |
| Per TB/s · hour (median) | $1.001 | $0.483 | +107% |
| Trend | |||
| 30-day median change | +18.2% | +9.2% | — |
| 90-day median change | +18.2% | +9.2% | — |
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
Airon AI1× $1.88 - 2
Vast.aiTrust 784× $2.53 - 3
GPURento1× $2.59
- 1
Cyfuture AI8× $1.04 - 2
io.netTrust 428× $1.16 - 3
Sesterce CloudTrust 608× $1.21
Verdict · computed from the numbers aboveChoose one, choose the other
Choose the H100 NVL if…
- you want choice: 13 providers have a fresh offer today against 4.
Choose the Gaudi 2 if…
- your model needs more than 94 GB per GPU: 96 GB against 94 GB.
- the hourly price matters most: median $1.18 against $3.94, 70% less.
- you serve LLMs: memory bandwidth per dollar is better ($0.483 per TB/s-hour against $1.001).
Questions about H100 NVL vs Gaudi 2
Which is cheaper to rent, the H100 NVL or the Gaudi 2?
Today the cheapest current on-demand H100 NVL is $1.88 per GPU-hour and the cheapest Gaudi 2 is $1.04; medians are $3.94 and $1.18. Cheapest excludes offers marked unavailable and counts one floor per marketplace.
How much memory do the H100 NVL and Gaudi 2 have?
94 GB of HBM3 on the H100 NVL, 96 GB of HBM2E on the Gaudi 2, at 3.94 and 2.45 TB/s.
Where do these numbers come from?
From 36 current offers read on the providers' own pages and APIs over the last 45 days (reference 00:18 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.