A100 40GB vs H100 SXM

Open Advisor →

Side-by-side GPU comparison: specs, memory, compute performance, and live cloud pricing.

Verdict

H100 SXM has 40 GB more VRAM, making it better suited for large models and long context windows. For compute-bound workloads like training, H100 SXM delivers 3.2× higher FP16 throughput. At $0.93/hr vs $1.80/hr, A100 40GB is the more cost-efficient choice for inference. H100 SXM supports a broader range of models (636 vs 612 from this catalog), giving more flexibility.

Specifications

A100 40GBH100 SXM
VRAM40 GB80 GB
VRAM TypeHBM2HBM3
Memory Bandwidth1.6 TB/s3.4 TB/s
FP16 Performance312 TFLOPS990 TFLOPS
ManufacturerNVIDIANVIDIA
FP8 SupportNoYes
FP4 SupportNoNo

Price / Performance

Based on cheapest single-GPU on-demand pricing. Lower $/TFLOP = better compute value; lower $/GB = better memory value.

A100 40GBH100 SXM
$/hr (cheapest)$0.93✓ best$1.80
$/TFLOP (compute value)$0.0030$0.0018✓ best
$/GB VRAM (memory value)$0.0232$0.0225✓ best

Cloud Pricing

Cheapest on-demand price per provider (single GPU).

A100 40GB

ProviderOn-demandSpotRent
Vast.ai$0.93/hrRent
Google Cloud$1.61/hr$1.17/hr
Lambda$1.99/hr

H100 SXM

ProviderOn-demandSpotRent
Vast.ai$1.80/hrRent
RunPod$2.39/hr$1.25/hrRent
Nebius$2.95/hr$1.25/hr
Lambda$3.29/hr
Google Cloud$4.20/hr$1.14/hr
Amazon Web Services$6.88/hr$2.81/hr
Microsoft Azure$6.98/hr$6.98/hr

Model Compatibility

Models from the catalog that fit on each GPU, grouped by required precision.

A100 40GB (612 models)

H100 SXM (636 models)

You might also compare…

Pricing data refreshed hourly · Last updated April 11, 2026 · Browse all comparisons