H200 SXM 141GB vs H100 SXM5 80GB

Compare NVIDIA H200 SXM 141GB and NVIDIA H100 SXM5 80GB specs, performance, and cloud pricing

H200 SXM 141GB

141GB

From $3.49/hr

H100 SXM5 80GB

80GB

From $2.20/hr

Architecture

Hopper

vs Hopper

FP16 Gap

2.0x

H100 SXM5 80GB leads

SpecificationH200 SXM 141GBH100 SXM5 80GB
VRAM141 GB80 GB
VRAM TypeHBM3eHBM3
FP16 TFLOPS989.5 TFLOPS2.0 PFLOPS
FP8 TFLOPS2.0 PFLOPS4.0 PFLOPS
Memory Bandwidth4.8 TB/s3.4 TB/s
TDP700W700W
InterconnectNVLink 4NVLink 4
ArchitectureHopperHopper

Price Comparison

MetricH200 SXM 141GBH100 SXM5 80GB
Cheapest On-Demand$3.49/hr$2.20/hr
Cheapest Spot$2.52/hr$1.35/hr
Providers Available47

Verdict

Best for Training

NVIDIA H100 SXM5 80GB

2.0 PFLOPS FP16 with 80GB VRAM

Best Value

NVIDIA H100 SXM5 80GB

900 TFLOPS per $/hr

Best for Inference

NVIDIA H100 SXM5 80GB

4.0 PFLOPS FP8/FP16

Use-Case Recommendations

Large-Scale Training

Training LLMs and large multi-modal models

Winner

H100 SXM5 80GB

2.0 PFLOPS FP16 with 80GB HBM3 provides the best training throughput.

Inference at Scale

Deploying models in production for real-time inference

Winner

H100 SXM5 80GB

4.0 PFLOPS FP8/FP16 gives superior inference throughput.

Budget-Conscious Workloads

Getting the best performance per dollar

Winner

H100 SXM5 80GB

Starting at $2.20/hr delivers the best TFLOPS per dollar.

Learn More