NVIDIA H100 NVL 94GB vs NVIDIA GH200 96GB HBM3

The GH200 delivers 12% more FP32 throughput than the H100 (67 vs 60 TFLOPS dense). The GH200 also carries 2 GB more memory (96 GB vs 94 GB).

H100: Hopper, 2023 GH200: Hopper, 2023
FP32 dense lead
+12%
GH200: 67 vs 60 TFLOPS
Memory
94 vs 96 GB
2 GB more for the GH200
Bandwidth
3.9 TB/s vs 4.0 TB/s
GH200 moves data faster
TDP
400 W vs 1.0 kW
H100 draws 600 W less

Specifications

H100
GH200
Architecture
HopperHopper
Launch Year
20232023
Form Factor
PCIeSuperchip
Memory
94 GB96 GB
Memory Bandwidth
3.9 TB/s4.0 TB/s
TDP
400 W1.0 kW
Process Node
4nm4nm

Performance (TFLOPS)

H100
GH200
FP64
30 TFLOPS 34 TFLOPS
FP32
60 TFLOPS 67 TFLOPS
TF32
The vendor publishes only the 2:4 sparse figure for this precision
835 TFLOPS sparse
494 TFLOPS
989 TFLOPS sparse
BF16
The vendor publishes only the 2:4 sparse figure for this precision
1,671 TFLOPS sparse
990 TFLOPS
1,979 TFLOPS sparse
FP16
The vendor publishes only the 2:4 sparse figure for this precision
1,671 TFLOPS sparse
990 TFLOPS
1,979 TFLOPS sparse
FP8
The vendor publishes only the 2:4 sparse figure for this precision
3,341 TFLOPS sparse
1,979 TFLOPS
3,958 TFLOPS sparse
FP6
No verified data available No verified data available
FP4
No verified data available No verified data available
INT8
The vendor publishes only the 2:4 sparse figure for this precision
3,341 TOPS sparse
1,979 TOPS
3,958 TOPS sparse
Tensor Core
FP64 (TC)
60 TFLOPS 67 TFLOPS

FLOPS by Precision

What actually differs

Both GPUs launched in 2023: the H100 on Hopper and the GH200 on Hopper.

Dense throughput for the H100 against the GH200: FP64 30 vs 34 TFLOPS, FP32 60 vs 67 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.

Memory is 94 GB against 96 GB, fed at 3.9 TB/s versus 4.0 TB/s. More memory per GPU means larger models fit before you have to shard across cards, which often matters more than raw TFLOPS for inference.

Power budgets are 400 W for the H100 and 1.0 kW for the GH200. At FP32 that works out to 0.15 against 0.07 TFLOPS per watt.

For LLM training and inference, weight the FP8 and FP16 rows and memory capacity most heavily. For scientific and HPC workloads, the FP64 row is the one to read.

Frequently asked questions

Is the H100 faster than the GH200?

At FP32 precision the GH200 reaches 67 TFLOPS dense against 60 TFLOPS for the H100. The performance table on this page lists every published precision for both GPUs.

Which has more memory, the H100 or the GH200?

The GH200 carries 96 GB of memory versus 94 GB for the H100. Memory bandwidth is 3.9 TB/s for the H100 and 4.0 TB/s for the GH200.

How much power do the H100 and the GH200 draw?

The H100 is rated at 400 W TDP and the GH200 at 1.0 kW. On FP32 throughput per watt, the H100 is the more efficient part.

Can I rent the H100 or the GH200 in the cloud?

Yes. Live cloud listings tracked by Flopper start at $2.29 per GPU hour. The rental pricing section on this page lists current providers and rates for both GPUs.

Where to Rent

NVIDIA H100 NVL 94GB

ProviderConfigurationPrice/GPU-hrChecked
RunPod
1× H100 NVL Community $2.59 4h ago View →
Vast.ai
1× H100 NVL Community $2.67 4h ago View →
RunPod
1× H100 NVL $3.19 4h ago View →
Microsoft Azure
1× H100 NVL $6.98 4h ago View →
Microsoft Azure
2× H100 NVL $6.98 4h ago View →
Microsoft Azure
1× H100 NVL $6.98 4h ago View →
All H100 listings and price history →

NVIDIA GH200 96GB HBM3

ProviderConfigurationPrice/GPU-hrChecked
Lambda Labs
1× GH200 96GB HBM3 $2.29 4h ago View →
CoreWeave
1× GH200 96GB HBM3 $6.50 4h ago View →
All GH200 listings and price history →

Get Comparison Updates

New GPUs added weekly. Be the first to see how they compare.