Two Nvidia data center GPUs on the Hopper architecture: 80 GB HBM3 against 141 GB HBM3e and 43% more memory bandwidth on the H200.
At a glance
Updated 6 minutes ago- Memory
- 80 GB
- HBM3 · 94 GB on NVL
- Memory bandwidth
- 3.35 TB/s
- 2,000 GB/s on PCIe · 3,938 GB/s on NVL
- From
- $1.50 /GPU/hr
- 53 providers
- Memory
- 141 GB
- HBM3e
- Memory bandwidth
- 4.80 TB/s
- From
- $2.09 /GPU/hr
- 46 providers
Price comparison
Cheapest listed price per GPU per hour by billing type, and the median across providers.
| Billing | H100 | H200 | Difference |
|---|---|---|---|
| On-demand | $1.50 | $2.09 | 28% cheaper on H100 |
| Spot | $0.17 | $2.45 | 14x more expensive on H200 |
| Reserved | $1.53 | $2.29 | 33% cheaper on H100 |
| Median | $3.39 | $4.40 | 23% cheaper on H100 |
Median: Median across providers, one vote each, so a large catalog does not outweigh a small one.
Price history
H100 vs H200 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | H100 | H200 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 80 GB HBM3 94 GB on NVL | 141 GB HBM3e | 1.8x more on H200 |
| Bandwidth | 3,350 GB/s 2,000 GB/s on PCIe 3,938 GB/s on NVL | 4,800 GB/s | 1.4x higher on H200 |
| Bus width | 5,120-bit | ||
| Compute · dense | |||
| FP8 | 1,979 TFLOPS | 1,979 TFLOPS | Identical |
| FP16 / BF16 | 989.5 TFLOPS | 989.5 TFLOPS | Identical |
| INT8 | 1,979 TOPS | 1,979 TOPS | Identical |
| FP32 | 67 TFLOPS | 67 TFLOPS | Identical |
| FP64 | 34 TFLOPS | 34 TFLOPS | Identical |
| Platform | |||
| Interconnect | NVLink 900 GB/s NVLink 600 GB/s on PCIe, NVL | NVLink 900 GB/s | Same headline SKU |
| Board power | 700 W | 700 W | Identical |
| Released | Q3 2022 | Q4 2023 | 14 months newer on H200 |
| Precision | |||
| Supported formats | FP8FP16BF16TF32FP32FP64INT8 | FP8FP16BF16TF32FP32FP64INT8 | |
Figures are for each model's headline configuration. Where a form factor carries a different number, it is named under the figure.
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: H100 datasheet and H200 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | H100 /GPU/hr | H200 /GPU/hr |
|---|---|---|
Lium
|
On-Demand $1.50 | On-Demand $3.00 |
Koyeb
|
On-Demand $2.50 | On-Demand $3.00 |
GPU.ai
|
On-Demand $1.98 | On-Demand $3.29 |
Vast.ai
|
On-Demand $2.67 | On-Demand $3.09 |
Massed Compute
|
On-Demand $2.73 | On-Demand $3.62 |
Runcrate
|
On-Demand $2.75 | On-Demand $4.40 |
HyperAI
|
On-Demand $1.80 | -- |
Spheron
|
On-Demand $2.65 | -- |
Verda
|
-- | On-Demand $4.00 |
Lyceum
|
-- | On-Demand $4.29 |
53 providers list the H100 and 46 the H200. Every listing is on the H100 page and the H200 page.
Similar GPUs
Other comparisons in the same class.
Our data for Nvidia H100 was last updated on Sept. 7, 2026, and for Nvidia H200 on Sept. 7, 2026.