NVIDIA GH200 96GB HBM3 vs NVIDIA H100 NVL 94GB

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).

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

Specifications

GH200
H100
Architecture
HopperHopper
Launch Year
20232023
Form Factor
SXMPCIe
VRAM
96 GB94 GB
Memory Bandwidth
4.0 TB/s3.9 TB/s
TDP
1.0 kW400 W
Process Node
4nm4nm

Performance (TFLOPS)

GH200
H100
FP64
34.0 TFlops 30.0 TFlops
FP32
67.0 TFlops 60.0 TFlops
FP16
990.0 TFlops
1979.0 TFLOPS sparse
BF16
990.0 TFlops
1979.0 TFLOPS sparse
FP8
1979.0 TFlops
3958.0 TFLOPS sparse
INT8
1979.0 TFlops
3958.0 TOPS sparse
Tensor Core
FP64 (TC)
60.0 TFlops

FLOPS by Precision

What actually differs

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

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

Memory is 96 GB against 94 GB, fed at 4.0 TB/s versus 3.9 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 1.0 kW for the GH200 and 400 W for the H100. At FP32 that works out to 0.07 against 0.15 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 GH200 faster than the H100?

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 GH200 or the H100?

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

How much power do the GH200 and the H100 draw?

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

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

Yes. Live cloud listings tracked by Flopper start at $2.59 per GPU hour. The rental pricing section on this page lists current providers and rates for both GPUs.

Where to Rent

NVIDIA GH200 96GB HBM3

ProviderConfigurationPrice/GPU-hrChecked
CoreWeave
1× GH200 96GB HBM3 $6.50 2h ago View →
All GH200 listings and price history →

NVIDIA H100 NVL 94GB

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

Get Comparison Updates

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