The table below compares current hourly H100 pricing for a single NVIDIA H100 80GB GPU across major U.S. cloud providers.
To help you compare Nvidia H100 prices across clouds, this guide normalizes per-GPU rates (even when only multi-GPU instances are listed) and uses each provider's currently published U.S. pricing references.
| Provider | SKU/Instance | $/GPU-hr* | Notes |
|---|---|---|---|
| Thunder Compute | H100 | $2.19 | Lowest fixed-price on-demand rate in this comparison. |
| Vultr | H100 (8 x H100) | $2.30 | 24-month commitment price, not on-demand; normalized from an 8-GPU node. |
| Vast.ai | H100 (third-party hosts) | $2.01 | Marketplace current price per H100 GPU. Reliability varies. |
| Hyperbolic | H100 PCIe | $2.69 | Marketplace-style H100 pricing. |
| CoreWeave | NVIDIA HGX H100 (8 x H100) | $2.70 | $21.60/hr node price, normalized per GPU. |
| RunPod | H100 PCIe | $2.89 | Public listed H100 price; availability and reliability can vary. |
| Nebius | H100 PCIe | $3.85 | Public listed single-GPU equivalent. |
| Crusoe Cloud | H100 PCIe | $3.90 | Public listed hourly pricing. |
| Modal | H100 | $3.95 | Public listed H100 rate. |
| Lambda | H100 PCIe (1x) | $3.99 | Public listed 1-GPU rate on Lambda pricing page. |
| Paperspace | H100 (Dedicated instance) | $5.95 | Public listed price. |
| CoreWeave | NVIDIA HGX H100 (8 x H100) | $6.16 | $49.24/hr node price, normalized per GPU. |
| AWS | p5.4xlarge | $6.88 | p5.4xlarge pricing in U.S. regions; avoid comparing against reserved Capacity Blocks rates. |
| Azure | NC40ads_H100_v5 VM | $8.30 | Single H100 GPU VM in East US region. |
| Oracle Cloud | BM.GPU.H100.8 | $10.00 | 8-GPU bare-metal node: $80.00/hr. |
| Google Cloud | a3-highgpu-1g | $11.06 | 1x H100 80GB in US-central (on-demand). |
| Normalized cost per single H100 GPU, even when provider only offers multi-GPU instances. Last update: July 1, 2026. |
|||
Methodology (why you can trust these numbers)
- Published list rates: Provider pages only, with normalized per-GPU pricing. Custom quote and commitment-only listings are labeled.
- Same class of silicon: All prices refer to NVIDIA H100 80GB GPUs.
- Normalized across SKUs: Where providers list only 8-GPU nodes, we divide by eight to get a single-GPU equivalent.
- USD in U.S. regions: Rates elsewhere can differ by 5-20 %.
Why this matters for developers
This table compares how much it costs to run an NVIDIA H100 for 10 hours across major cloud providers.
| Provider | Listed $/GPU-hr | 10-Hour Cost |
|---|---|---|
| Vast.ai | $2.01 | $20.10 |
| Thunder Compute | $2.19 | $21.90 |
| Vultr | $2.30 | $23.00 |
| Hyperbolic | $2.69 | $26.90 |
| RunPod | $2.89 | $28.90 |
| Nebius | $3.85 | $38.50 |
| Crusoe Cloud | $3.90 | $39.00 |
| Modal | $3.95 | $39.50 |
| Lambda | $3.99 | $39.90 |
| Paperspace | $5.95 | $59.50 |
| CoreWeave | $6.16 | $61.60 |
| AWS | $6.88 | $68.80 |
| Azure | $8.30 | $83.00 |
| Oracle Cloud | $10.00 | $100.00 |
| Google Cloud | $11.06 | $110.60 |
H100 availability is extremely low across cloud providers, so consider using an A100 instead when your workload does not strictly require H100 performance.
| Provider | GPU | Listed $/GPU-hr | 10-Hour Cost |
|---|---|---|---|
| Thunder Compute | A100 | $1.09 | $10.90 |
Price sources: provider pricing pages linked in the table above. Vast.ai values are marketplace estimates and vary by host and region.
H100 workloads are expensive by default, so even small hourly price differences have an outsized impact. Whether you're training large models or running high-throughput inference, choosing a lower-cost provider can dramatically extend your budget.
Takeaways
- Thunder Compute's H100 rate is about 3x cheaper than AWS, about 4x cheaper than Azure, and about 5x cheaper than GCP.
- Specialized providers like RunPod and Lambda are closer than hyperscalers, while Vast.ai can list lower marketplace H100 rates with more host-level variability.
- Bookmark this page, we refresh the numbers quarterly so you do not have to.
- Building a startup? See our analysis of Startup-Friendly GPU Cloud Providers for credit offers, and spin up an H100 on Thunder Compute.
Read a full comparison of AWS P5 instances powered by NVIDIA H100 and H200 GPUs.
Compare other GPUs and maximize your computing power by reading the cheapest cloud GPU providers.