gpu catalog: h100 nvl
The inference-tuned H100 variant: 94 GB of HBM3 and higher bandwidth than the SXM part, purpose-built for serving LLMs in PCIe servers.
| GPU | VRAM | Regions | Availability | Price / GPU | |
|---|---|---|---|---|---|
| H100 NVL | 94G | 3 regions | available | $3.19/hr | launch |
The inference-tuned H100 variant: 94 GB of HBM3 and higher bandwidth than the SXM part, purpose-built for serving LLMs in PCIe servers. VRAM94 GB HBM3 Memory bandwidth3.9 TB/s FP16 tensor835 TFLOPS ArchitectureHopper InterconnectPCIe 5.0 + NVLink bridge Board power400 W Released2023 Regionsus-central, us-east, us-west | |||||
| H100 NVLcommunity | 94G | 2 regions | running low | $2.37/hr | launch |
| L40S | 48G | 4 regions | high availability | $0.80/hr | launch |
| RTX PRO 6000 | 96G | 4 regions | high availability | $1.99/hr | launch |
| RTX 6000 ADA | 48G | 2 regions | high availability | $0.65/hr | launch |
| RTX 5090community | 32G | 8 regions | high availability | $0.35/hr | launch |
| L40 | 48G | 2 regions | high availability | $0.78/hr | launch |
| RTX 4090community | 24G | 8 regions | high availability | $0.27/hr | launch |
| RTX A6000 | 48G | 2 regions | high availability | $0.45/hr | launch |
| A100 | 80G | 6 regions | high availability | $1.07/hr | launch |
| RTX PRO 4500 | 32G | 2 regions | high availability | $0.72/hr | launch |
| H100 SXM | 80G | 5 regions | high availability | $3.49/hr | launch |
| A100community | 80G | 5 regions | high availability | $0.88/hr | launch |
| RTX 3090community | 24G | 6 regions | high availability | $0.12/hr | launch |
| H200 SXM | 141G | 6 regions | high availability | $3.59/hr | launch |
| RTX 4090 | 24G | 3 regions | high availability | $0.74/hr | launch |
| L40Scommunity | 48G | 4 regions | available | $0.79/hr | launch |
| H100 PCIE | 80G | 2 regions | available | $1.98/hr | launch |
| RTX A4000community | 16G | 5 regions | available | $0.08/hr | launch |
| RTX 5080community | 16G | 7 regions | available | $0.16/hr | launch |
| A100 40GBcommunity | 40G | 6 regions | available | $0.56/hr | launch |
| RTX 5090 | 32G | 2 regions | available | $0.99/hr | launch |
| L4 | 24G | 4 regions | available | $0.44/hr | launch |
| RTX 3080community | 10G | 5 regions | available | $0.10/hr | launch |
| H200 NVLcommunity | 141G | 3 regions | available | $3.98/hr | launch |
| RTX 6000 ADAcommunity | 48G | 2 regions | available | $0.74/hr | launch |
| B300 | 288G | 2 regions | available | $7.89/hr | launch |
| H100 PCIEcommunity | 80G | 2 regions | available | $2.54/hr | launch |
| H200 SXMcommunity | 141G | 4 regions | available | $3.98/hr | launch |
| H100 SXMcommunity | 80G | 3 regions | running low | $2.27/hr | launch |
| A40 | 48G | 2 regions | running low | $0.49/hr | launch |
| RTX 3090 | 24G | 2 regions | running low | $0.50/hr | launch |
| B200 | 192G | 2 regions | running low | $6.79/hr | launch |
| RTX A5000community | 24G | 3 regions | running low | $0.23/hr | launch |
| L4community | 24G | 2 regions | running low | $0.27/hr | launch |
| B200community | 192G | 2 regions | running low | $5.32/hr | launch |
| RTX A6000community | 48G | 2 regions | running low | $0.41/hr | launch |
| RTX 4080community | 16G | 3 regions | running low | $0.18/hr | launch |
| RTX PRO 6000community | 96G | 1 region | running low | $1.69/hr | launch |
| A100 40GB | 40G | 1 region | running low | $0.89/hr | launch |
| V100community | 16G | 2 regions | running low | $0.09/hr | launch |
lowest live on-demand rate per GPU, refreshed continuously · per-second billing
| Provider | Price / GPU / hr | Notes |
|---|---|---|
| GPU.ai (Secure Cloud) | $3.19/hr | Vetted data center capacity, per-second billing |
| GPU.ai (Community Cloud) | $2.37/hr | Marketplace capacity from independent suppliers |
| CoreWeave | — | Public on-demand list rate |
| AWS | — | Public on-demand list rate |
| Azure | — | Public on-demand list rate |
competitor rates are public on-demand list prices, refreshed daily · gpu.ai prices are live floor rates with zero markup
faq
H100 NVL cloud instances on GPU.ai currently start at $2.37 per GPU per hour, billed per second. That is the live floor price across every provider in our fleet. We route your launch to the cheapest provider with stock and add zero markup.
The NVIDIA H100 NVL has 94 GB of HBM3 memory with up to 3,900 GB/s of bandwidth. GPU.ai reports the minimum VRAM actually delivered across our fleet, so the figure you see at launch is the figure you get.
GPU.ai aggregates a dozen-plus GPU clouds and passes through each provider's raw price with no markup, so the same silicon is typically far cheaper than hyperscaler list rates. The comparison table on this page shows the current public on-demand list prices side by side.
Billing is per second, starting when your instance is SSH-ready and stopping the moment you terminate. There are no minimum commitments, no reservation fees, and no egress surprises: you pay the listed hourly rate pro-rated to the second.