Side-by-side GPU comparison: specs, memory, compute performance, and live cloud pricing.
Verdict
For compute-bound workloads like training, H100 PCIe delivers 2.4× higher FP16 throughput. At $1.39/hr vs $1.80/hr, A100 80GB is the more cost-efficient choice for inference.
Specifications
| A100 80GB | H100 PCIe | |
|---|---|---|
| VRAM | 80 GB | 80 GB |
| VRAM Type | HBM2e | HBM2e |
| Memory Bandwidth | 2.0 TB/s | 2.0 TB/s |
| FP16 Performance | 312 TFLOPS | 757 TFLOPS |
| Manufacturer | NVIDIA | NVIDIA |
| FP8 Support | No | Yes |
| FP4 Support | No | No |
Price / Performance
Based on cheapest single-GPU on-demand pricing. Lower $/TFLOP = better compute value; lower $/GB = better memory value.
| A100 80GB | H100 PCIe | |
|---|---|---|
| $/hr (cheapest) | $1.39✓ best | $1.80 |
| $/TFLOP (compute value) | $0.0045 | $0.0024✓ best |
| $/GB VRAM (memory value) | $0.0174✓ best | $0.0225 |
Cloud Pricing
Cheapest on-demand price per provider (single GPU).
Model Compatibility
Models from the catalog that fit on each GPU, grouped by required precision.
A100 80GB (636 models)
H100 PCIe (636 models)
You might also compare…
Pricing data refreshed hourly · Last updated April 11, 2026 · Browse all comparisons