Specifications Compared
| Spec | H100 | P100 |
|---|---|---|
| TDP | 700W | 250W |
| VRAM | 80-94 GB | 16 GB |
| CUDA Cores | 16,896 | 3,584 |
| FP8 (dense) | 1,979 TFLOPS | Not published |
| Memory Type | HBM3 | HBM2 |
| Architecture | Hopper | Pascal |
| FP16 (dense) | 989 TFLOPS | 18.7 TFLOPS |
| Form Factors | SXM5, PCIe, NVL | SXM2, PCIe |
| INT8 (dense) | 1,979 TOPS | Not published |
| Interconnect | NVLink, PCIe 5.0, InfiniBand | NVLink, PCIe 3.0 |
| Tensor Cores | 528 | Not published |
| FP32 Performance | 67 TFLOPS | 9.3 TFLOPS |
| FP64 Performance | 34 TFLOPS | 4.7 TFLOPS |
| Memory Bandwidth | 3,350 GB/s | 732 GB/s |
| FP8 (with sparsity) | 3,958 TFLOPS | Not published |
| FP16 (with sparsity) | 1,979 TFLOPS | Not published |
| INT8 (with sparsity) | 3,958 TOPS | Not published |
Performance Analysis
Dense FP16 performance on the H100 measures 989 TFLOPS while the P100 reaches 18.7 TFLOPS. The H100 also delivers 1979 TFLOPS with sparsity in FP16 compared to the P100 dense figure of 18.7 TFLOPS. Such deltas allow the H100 to process larger models during training and inference at higher throughput rates than the P100. Memory bandwidth of 3350 GB/s on the H100 versus 732 GB/s on the P100 supports increased batch sizes without memory constraints. FP32 throughput at 67 TFLOPS on the H100 exceeds the 9.3 TFLOPS on the P100 by a substantial margin in precision sensitive computations.
Current On-Demand Offers
Cheapest secure on-demand offer per provider that is reported in stock right now, per GPU per hour; multi-GPU instances show the whole-instance price alongside. Deploy opens the DeployGPU console for providers on the platform, otherwise the provider's own site. Stock is rechecked every minute.
H100
| Provider | Region | GPUs | Per GPU / hr | Instance / hr | Deploy |
|---|---|---|---|---|---|
| Hyperstack | CANADA-1 | 1 | $2.50 | — | Deploy |
| Vast.ai | Czechia, CZ | 2 | $2.59 | $5.17 | Deploy |
| QuantaCloud | us-midwest-2 | 1 | $2.59 | — | Deploy |
| Massed Compute | us-central-3 | 2 | $2.73 | $5.46 | Deploy |
| Ori | dallas-2 | 2 | $2.90 | $5.80 | Deploy |
9 providers in stock, 25 offers (cheapest per provider shown). All H100 offers, price history and alerts
P100
1 provider in stock, 2 offers (cheapest per provider shown). All P100 offers, price history and alerts
Notify me when H100 drops below a price
One email when the cheapest in-stock on-demand price per GPU-hour falls below your threshold. Cheapest right now: $2.50/GPU-hr.
QuantaCloud
Comparing H-series providers? We broker across all of them.
Hopper stock changes by the hour and prices differ by provider. If you need 16+ GPUs reserved or a cluster in the next 90 days, we quote H-series or B300 inventory at partner rates: one quote, 24h turnaround.
When to Choose the H100
Its 80 GB to 94 GB memory capacity and 3350 GB/s bandwidth enable handling of extensive datasets in scientific computing tasks. Organizations running inference on models that demand 1979 TFLOPS dense FP8 or 3958 TFLOPS with sparsity select the H100 for its architectural advantages over prior generations.
When to Choose the P100
The P100 fits smaller scale scientific computing workloads that operate within 16 GB memory limits and require only 9.3 TFLOPS FP32. Legacy systems already configured for Pascal architecture continue to use the P100 for FP16 dense tasks at 18.7 TFLOPS where migration costs outweigh benefits. Budget constrained environments with modest batch size needs may retain the P100 despite lower 732 GB/s bandwidth.
Use Cases
The H100 supplies 989 TFLOPS dense FP16 and up to 94 GB memory to accommodate large models.
The H100 achieves 1979 TFLOPS dense FP8 for faster token generation than the P100 18.7 TFLOPS FP16.
The H100 3350 GB/s bandwidth permits larger batch sizes during adaptation of models compared to the P100.
The H100 1979 TFLOPS with sparsity in FP16 accelerates image generation workloads beyond P100 capabilities.
The P100 handles smaller FP32 jobs at 9.3 TFLOPS while the H100 scales to 67 TFLOPS for demanding simulations.
Frequently Asked Questions
How does FP16 performance differ between the H100 and P100?▾
The H100 provides 989 TFLOPS dense FP16 and 1979 TFLOPS with sparsity. The P100 offers 18.7 TFLOPS dense FP16.
What memory configurations distinguish the H100 from the P100?▾
The H100 contains between 80 GB and 94 GB HBM3 memory. The P100 holds 16 GB HBM2 memory.
Which GPU offers higher memory bandwidth?▾
The H100 reaches 3350 GB/s bandwidth. The P100 reaches 732 GB/s bandwidth.
How do FP32 figures compare on the H100 and P100?▾
The H100 delivers 67 TFLOPS FP32. The P100 delivers 9.3 TFLOPS FP32.
What TDP values apply to each GPU?▾
The H100 lists a TDP of 700W. The P100 lists a TDP of 250W.
Which interconnect options appear on the H100 versus the P100?▾
The H100 supports NVLink, PCIe 5.0 and InfiniBand. The P100 supports NVLink and PCIe 3.0.
Which is cheaper to rent, the H100 or the P100?▾
Cloud rental prices for both the H100 and P100 vary by provider, configuration, and availability. This page shows live pricing from 25+ providers updated every 60 seconds. Scroll to the Live Cloud Pricing section to compare current rates.
How much VRAM does the H100 have compared to the P100?▾
The H100 has 80 to 94 GB of HBM3 memory. The P100 has 16 GB of HBM2 memory.
Can I find H100 and P100 GPUs available to rent right now?▾
Yes. This page shows real-time availability across 25+ cloud GPU providers. The Live Cloud Pricing section displays only in-stock offers with current pricing.
What is the main difference between the H100 and the P100?▾
The H100 uses the Hopper architecture (2022) while the P100 uses Pascal (2016). The H100 delivers 52.9x the dense FP16 throughput (989 vs 18.7 TFLOPS, both without sparsity) and 4.6x the memory bandwidth of the P100.
Rent these GPUs
Each GPU page lists every current on-demand offer by provider, per GPU-hour, updated every minute.
Related comparisons
How this page is made
- Specifications come from the NVIDIA, AMD and Intel datasheets for the H100 and the P100. Dense and sparse throughput are listed separately.
- Prices and availability are live: every offer shown is an in-stock, on-demand listing from a provider we track, rechecked every minute.
- The written comparison (overview, performance notes, when to choose each card, verdict, use cases and FAQ) was drafted with an AI model from the specification table above and passed an automated check that rejects any figure not in that table. It was last generated on . It contains no prices; those are always read live.
- Read how we collect the data, or report an error on this page.
Next steps
- Rent H100Every current offer by provider, daily price history and a price alert.
- Rent P100Every current offer by provider, daily price history and a price alert.
- GPU price indexHow on-demand prices for the major GPUs have moved, updated daily.
- All GPUsEvery GPU we track, with current lows and provider counts.