H100NVL
94GB 400W 3 providers 136 offersNVIDIA H100 NVL is a 94 GB HBM3 PCIe Hopper GPU tuned for large language model inference, with roughly 3.9 TB/s of memory bandwidth and configurable 350-400 W power.
Cheapest
$1.227/hr
Vast.ai
Median
$8.373/hr
across 136 offers
P90
$18.148/hr
90th percentile
Memory
94 GB
TDP
400 W
Bandwidth
3900 GB/s
FP16
418 TFLOPS
Brand
NVIDIA
Series
NVIDIA
Year
2023
Arch
Hopper
Price by provider
cheapest rate · $/GPU/hr · lower is betterVast.ai
0 regions
$1.227/hr
Azure
0 regions
$1.509/hr
RunPod
0 regions
$2.590/hr
3 providers
Best Deal Right Now
VA
Vast.ai
$1.227/hr
per GPU / hour
Configurationx1 · H100NVL
System RAM24-321GB
Region—
Estimated Running Cost
At cheapest rate · per GPU
1 hour$1.23
24 hours$29.44
1 week$206.08
1 month$21491.26
Explore NVIDIA H100 NVL
Related comparisons, provider pages, raw specs, and model-fit tools.
NVIDIA H100 NVL vs NVIDIA H100Compare live rental prices, VRAM, FP16 throughput, and provider availability.NVIDIA H100 NVL vs NVIDIA H200Compare live rental prices, VRAM, FP16 throughput, and provider availability.NVIDIA H100 NVL vs NVIDIA A100Compare live rental prices, VRAM, FP16 throughput, and provider availability.NVIDIA H100 NVL on Vast.aiSee SKUs, regions, spot status, and current prices from $1.227/hr.NVIDIA H100 NVL on AzureSee SKUs, regions, spot status, and current prices from $1.509/hr.NVIDIA H100 NVL on RunPodSee SKUs, regions, spot status, and current prices from $2.590/hr.NVIDIA H100 NVL full specificationsOpen the raw GPU Ark spec detail page for clocks, memory, cores, and API fields.Find models that fit this class of GPUMatch LLM size and precision to GPUs with enough VRAM.