RunPod Community CloudIn stock
On-Demand from $0.50
From
$0.50 / GPU / hr
On-DemandVisit website →
Rental price index
NVIDIA H200 NVL
🏢PCIe H200 for air-cooled servers.
Prices collected 2026-09-12 · 5 providers, 23 configurations
Compare vs other GPUs →On-demand
Index median price per GPU per hour, by billing type. Nothing is interpolated.
At a glance
Median on-demand
$3.72/GPU/hr
Cheapest on-demand
$0.50RunPod Community Cloud
RunPod Community CloudFloor, any billing
$0.50incl. spot
Per month at 720h
$2,678
NVIDIA H200 NVL pricing by provider
One card per provider, cheapest first. A configuration is one priced unit — an instance type, a SKU, a marketplace offer — and the per-GPU price is that unit's price divided by its GPU count. Open a card to see every configuration with its vCPU, RAM and region where the provider publishes them.
5 providers · 23 configurations
GPU.aiIn stock
AEUAE2 configs1xcollected 2026-09-12
On-Demand from $3.29
From
$3.29 / GPU / hr
On-DemandVisit website →
Vast.aiIn stock
USUSA17 configs1x–7xcollected 2026-09-12
On-Demand from $3.34
From
$3.34 / GPU / hr
On-Demand · Montana, USVisit website →Massed ComputeIn stock
USUSA2 configs1x–4xcollected 2026-09-12
On-Demand from $3.62
From
$3.62 / GPU / hr
On-DemandVisit website →
RunPod Secure CloudIn stock
USUSA1 config1xcollected 2026-09-12
On-Demand from $3.79
From
$3.79 / GPU / hr
On-DemandVisit website →What fits on a NVIDIA H200 NVL
Weights + KV cache at 8K context + runtime overhead, against 127 GB usable of the 141 GB on the card. Capacity follows total parameters, even for mixture-of-experts models — any expert may be needed next. Where one card is not enough, the table says how many are, and prices the set at today's median.
| Model | Params | 4-bit | 8-bit | 16-bit |
|---|---|---|---|---|
| 2.8T104B active | 13× GPU$48.36/hr | 24× GPU$89.28/hr | 45× GPU$167.40/hr | |
| 552B16B active | 3× GPU$11.16/hr | 5× GPU$18.60/hr | 9× GPU$33.48/hr | |
 Llama 4 Maverick | 400B17B active | 2× GPU$7.44/hr | 4× GPU$14.88/hr | 7× GPU$26.04/hr |
 gpt-oss-120b | 117B5.1B active | ✓ 67.9 GB~1339 tok/s | ✓ 126.4 GB~709 tok/s | 2× GPU$7.44/hr |
| Llama 3.1 70B | 70B | ✓ 44.5 GB~98 tok/s | ✓ 79.5 GB~52 tok/s | 2× GPU$7.44/hr |
 Gemma 4 31B | 31B | ✓ 21.2 GB~220 tok/s | ✓ 36.7 GB~117 tok/s | ✓ 65.7 GB~62 tok/s |
 Qwen3 Coder 30B-A3B | 30.5B3.3B active | ✓ 19 GB~2069 tok/s | ✓ 34.3 GB~1095 tok/s | ✓ 62.9 GB~582 tok/s |
| Qwen3.8 27B | 27B | ✓ 18.7 GB~253 tok/s | ✓ 32.2 GB~134 tok/s | ✓ 57.6 GB~71 tok/s |
| gpt-oss-20b | 21B3.6B active | ✓ 13.7 GB~1896 tok/s | ✓ 24.2 GB~1004 tok/s | ✓ 43.9 GB~533 tok/s |
| Llama 3.1 8B | 8B | ✓ 6.9 GB~853 tok/s | ✓ 10.9 GB~452 tok/s | ✓ 18.4 GB~240 tok/s |
tok/s is decode throughput from memory bandwidth, derated 20%, single stream — an estimate, not a benchmark. Set cost uses today's median × GPUs required.
Specifications
Memory141 GB HBM3e
Memory bandwidth4,800 GB/s
ArchitectureHopper
VendorNVIDIA
ClassDatacenter
Released2024-11
Bandwidth is listed because it governs generation speed: each token streams the active weights once, so tokens per second is bandwidth divided by bytes read. TFLOPs decide prompt processing and training, not how fast text appears.
Alternatives
NVIDIA H200141 GB · $4.30/hr median · 16 providersNVIDIA B200180 GB · $6.39/hr median · 15 providersNVIDIA GH20096 GB · $3.02/hr median · 3 providers
Questions
How much does it cost to rent a NVIDIA H200 NVL?
As of 2026-09-12, the median on-demand price across 5 providers is $3.72 per GPU-hour, about $2,678 a month at 720 hours. The cheapest we saw was $0.50 at RunPod Community Cloud.
Why do prices for the same GPU vary so much?
Hyperscalers bundle CPU, RAM and network into an instance price and charge for the ecosystem around it. Neoclouds sell the GPU more directly. Marketplaces are independent hosts competing on price, with the trade-offs of shared, third-party hardware. The per-GPU figure is the instance price divided by GPU count, which is the only way to compare an 8-GPU node with a single-card listing.
Is the median the price I will pay?
No — it is the middle of the market. Each provider's own median is taken first so a provider with many instance sizes counts once, then the median across providers. Spot and reserved rates are excluded from it; they set the floor shown separately. Use it to judge whether a quote is high or low, then verify with the provider.
Which models can a NVIDIA H200 NVL run?
With 141 GB per GPU, the table above shows which reference models fit on one card at 4, 8 and 16-bit, using the same memory arithmetic as the rest of Runyard: weights plus KV cache plus runtime overhead, with capacity following total parameters even for mixture-of-experts models. Where a model needs more than one GPU it says how many and what that set costs at today's median.
Own the hardware instead? Work out what your card holds with the same arithmetic.
Check what fits →