Comparison · GPU vs GPU · read 21:35 UTCA100 40GB vs Tesla V100 32GB
Specifications, today's prices from 38 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 40GB | Tesla V100 32GB | Difference | |
|---|---|---|---|
| Memory & compute | |||
| VRAM | 40 GB | 32 GB | +25% |
| Memory bandwidth | 1.55 TB/s | 0.90 TB/s | +73% |
| FP16 tensor | 312 TFLOPS | 125 TFLOPS | +150% |
| FP32 | 19.5 TFLOPS | 15.7 TFLOPS | +24% |
| Power | 250 W | 300 W | -17% |
| Prices today | |||
| Cheapest on-demand | $0.58 | $0.55 | +6% |
| Median on-demand | $1.98 | $2.48 | -20% |
| Highest on-demand | $4.48 | $4.39 | +2% |
| Cheapest spot | $0.68 | $0.51 | +34% |
| Cheapest reserved | $0.72 | $0.38 | +90% |
| Current offers (n) | 56 | 63 | -11% |
| Providers with a fresh offer | 21 | 17 | +24% |
| Value | |||
| Per GB of VRAM · hour (cheapest) | $0.015 | $0.017 | -15% |
| Per GB of VRAM · hour (median) | $0.050 | $0.078 | -36% |
| Per PFLOP FP16 · hour (median) | $6.346 | $19.862 | -68% |
| Per TB/s · hour (median) | $1.273 | $2.759 | -54% |
| Trend | |||
| 30-day median change | -4.1% | 0.0% | — |
| 90-day median change | — | -45.2% | — |
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
- 1Edgevana8× $0.58
- 2AtmosCompute2× $0.73
- 3
JarvisLabsTrust 461× $0.89
- 1
LiumTrust 408× $0.55 - 2
Cyfuture AI8× $0.55 - 3
VoltageGPU8× $1.10
Verdict · computed from the numbers aboveChoose one, choose the other
Choose the A100 40GB if…
- your model needs more than 32 GB per GPU: 40 GB against 32 GB.
- the hourly price matters most: median $1.98 against $2.48, 20% less.
- you serve LLMs: memory bandwidth per dollar is better ($1.273 per TB/s-hour against $2.759).
- you want choice: 21 providers have a fresh offer today against 17.
Choose the Tesla V100 32GB if…
- No metric above favours it today.
Questions about A100 40GB vs Tesla V100 32GB
Which is cheaper to rent, the A100 40GB or the Tesla V100 32GB?
Today the cheapest current on-demand A100 40GB is $0.58 per GPU-hour and the cheapest Tesla V100 32GB is $0.55; medians are $1.98 and $2.48. Cheapest excludes offers marked unavailable and counts one floor per marketplace.
How much memory do the A100 40GB and Tesla V100 32GB have?
40 GB of HBM2e on the A100 40GB, 32 GB of memory on the Tesla V100 32GB, at 1.55 and 0.90 TB/s.
Where do these numbers come from?
From 119 current offers read on the providers' own pages and APIs over the last 45 days (reference 21:35 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.