H200 vs A100

A100 is 68% cheaper

Comparing the H200 and A100 for cloud rental? The H200 is available from 6 providers starting at $1.40/hr, while the A100 starts at $0.45/hr across 9 providers. Below is a live breakdown of pricing, VRAM, compute throughput, and where to rent each GPU.

H200

VRAM
141 GB
TFLOPS
495
Cheapest
$1.40/hr
Cheapest deal
Verda
$1.40/hr
per GPU-hour
SKU1H200.141S.44V
RegionFIN-01
TypeSpot
AzureNebiusOCIRunPodVast.aiVerda
See all H200 offers

A100

Best price
VRAM
80 GB
TFLOPS
312
Cheapest
$0.45/hr
Cheapest deal
Verda
$0.45/hr
per GPU-hour
SKU1A100.40S.22V
RegionFIN-01
TypeSpot
AWSAzureCudoGCPLambdaOCI
See all A100 offers

H200 vs A100 at a glance

Spec profile

H200A100
H200 vs A100 spec comparison radar chartVRAMFP16BandwidthValueProviders

Each axis is scaled to the stronger GPU — the outer edge is better.

Value for money

Compute / $
FP16 TFLOPS per $/hr
354
691
VRAM / $
GB of VRAM per $/hr
101 GB
177 GB
H200A100Higher is better value

Specs are manufacturer peak FP16 Tensor figures; real-world throughput varies by workload, precision, and software stack.

H200 vs A100 specs & pricing

SpecificationH200A100
Cheapest price
$1.40/hr
$0.45/hr
VRAM
141 GB
80 GB
FP16 TFLOPS
495
312
Memory bandwidth
4800 GB/s
2039 GB/s
Release year
2024
2020
Cloud providers
6
9

Which should you choose?

  • A100 is the cheaper option to rent — about 68% less per GPU-hour ($0.45/hr vs $1.40/hr).
  • H200 has more VRAM (141 GB), so it fits larger models and bigger batch sizes without sharding.
  • H200 offers higher FP16 throughput, training and serving faster on compute-bound workloads.

Provider availability

Only H200
NebiusVast.ai
Both GPUs
AzureOCIRunPodVerda
Only A100
AWSCudoGCPLambdaVultr

Frequently asked questions

Is the H200 or A100 cheaper to rent?

A100 is currently cheaper, from $0.45/hr versus $1.40/hr for the H200 — roughly 68% less per GPU-hour.

How much VRAM do the H200 and A100 have?

The H200 has 141 GB of VRAM and the A100 has 80 GB. More VRAM lets you serve larger models and longer context windows on a single GPU.

Which is faster, the H200 or A100?

On paper the H200 is faster, with 495 FP16 TFLOPS versus 312. Real-world speed also depends on memory bandwidth and software stack.

Which cloud providers offer both the H200 and A100?

Both GPUs are available from Azure, OCI, RunPod, Verda. PlanetGPU tracks live H200 and A100 prices across every major cloud.

Live prices come from public provider catalogs and can change quickly. How we collect prices.

Related GPU comparisons