Skip to content

Hopper H200 SXM vs Hopper H100 SXM5

Complete side-by-side comparison of specs, performance, memory, power efficiency, and pricing.

NVIDIA

Hopper H200 SXM

74

Spec Wins

NVIDIA

Hopper H100 SXM5

73

Detailed Specifications

SpecHopper H200 SXMHopper H100 SXM5
ArchitectureHopper Hopper
Memory141GB HBM3e 80GB HBM3
Memory Bandwidth4,800 GB/s 3,350 GB/s
FP16 TFLOPS1,979 1,979
FP8 TFLOPS3,958 3,958
BF16 TFLOPS1,979 1,979
INT8 TOPS3,958 3,958
TDP700W 700W
InterconnectNVLink 4.0 (900 GB/s) (900 GB/s) NVLink 4.0 (900 GB/s) (900 GB/s)
Perf Score74 73
EcosystemCUDA CUDA
Est. Price$30,000 $25,000

Hopper H200 SXM — Best For

LLM InferenceLarge Models

Hopper H100 SXM5 — Best For

LLM TrainingHPC

Who Should Choose Each GPU?

Choose Hopper H200 SXM if you…

  • Need maximum CUDA/TensorRT/vLLM ecosystem compatibility
  • Need more VRAM (141GB vs 80GB) for large model inference
  • Running LLM Inference workloads
  • Running Large Models workloads

Choose Hopper H100 SXM5 if you…

  • Need maximum CUDA/TensorRT/vLLM ecosystem compatibility
  • Working with a tighter CapEx budget (lower list price)
  • Running LLM Training workloads
  • Running HPC workloads

Verdict

The Hopper H200 SXM and Hopper H100 SXM5 target different priorities. The Hopper H200 SXM's 141GB of HBM3e gives it a clear edge for large-model inference where fitting the full model in VRAM eliminates quantization overhead. Both GPUs use CUDA, so ecosystem switching cost is not a factor. Use our TCO Calculator to model the full 3-year cost difference for your specific utilization and power costs.

Hopper H200 SXM vs Hopper H100 SXM5: Common Questions

Which is faster, Hopper H200 SXM or Hopper H100 SXM5?+

In FP8 throughput, the Hopper H100 SXM5 leads with 3,958 TFLOPS vs 3,958 TFLOPS. For LLM inference, memory capacity and bandwidth often matter more than raw TFLOPS — the Hopper H200 SXM has more VRAM (141GB).

Is Hopper H200 SXM or Hopper H100 SXM5 better for LLM training?+

For LLM training at scale, the Hopper H100 SXM5 has higher raw throughput. However, the choice also depends on your software stack: Hopper H200 SXM offers CUDA compatibility with the widest framework support (PyTorch, JAX, TensorRT).

What is the price difference between Hopper H200 SXM and Hopper H100 SXM5?+

The Hopper H200 SXM is estimated at $30,000 per unit and the Hopper H100 SXM5 at $25,000. Actual pricing varies by vendor, volume, and configuration. Check our Buy page for current reseller pricing.

Which GPU is more power efficient, Hopper H200 SXM or Hopper H100 SXM5?+

The Hopper H100 SXM5 has a lower TDP (700W vs 700W). Performance-per-watt depends on your workload — for FP8 inference, divide TFLOPS by TDP: Hopper H200 SXM = 5.7 TFLOPS/W vs Hopper H100 SXM5 = 5.7 TFLOPS/W.

Ask AI Advisor