H200 NVL offers
1 GPU × 1 hour · lowest estimate first
Listed GPU-hour rates for H200 NVL start at $3.32 an hour (on-demand) at IONOS Cloud.
Contract rates are hourly equivalents. A short run does not remove the minimum term; request a full contract quote.
| Provider | Region / listed unit | Job estimate | Offer |
|---|---|---|---|
| Per GPURegion not stated | $3.321 GPU billed · $3.32/GPU-h€3.00 listed · ECB 1.1052 on November 13, 2026 | View ↗ | |
| Per GPURegion not stated | $3.501 GPU billed · $3.50/GPU-h | View ↗ | |
| Per GPURegion not stated | $3.621 GPU billed · $3.62/GPU-h | View ↗ | |
| Per GPUSecure Cloud | $3.791 GPU billed · $3.79/GPU-h | View ↗ | |
| Per GPURegion not stated | $3.951 GPU billed · $3.95/GPU-h | View ↗ | |
| Per GPURegion not stated | $4.671 GPU billed · $4.67/GPU-h | View ↗ | |
| Per GPURegion not stated | $5.501 GPU billed · $5.50/GPU-h | View ↗ | |
| 8-GPU nodeRegion not stated | $43.748 GPUs billed · $5.47/GPU-h515.25 INR listed · ECB 0.010611 on November 13, 2026 | View ↗ |
| Provider | Region / listed unit | Job estimate | Offer |
|---|---|---|---|
| Per GPURegion not stated | $3.941 GPU billed · $3.94/GPU-h | View ↗ | |
| Per GPURegion not stated | $4.661 GPU billed · $4.66/GPU-h | View ↗ | |
| 2-GPU nodeMulti-GPU machines | $8.542 GPUs billed · $4.27/GPU-h | View ↗ |
No currently confirmed quotes for this rental type. Choose another rental type or GPU.
Estimate from listed rates. CPU, storage, transfer and tax may be extra. Live capacity is not verified.
As of , the cheapest listed rate for the NVIDIA H200 NVL is $3.32 per GPU-hour on IONOS Cloud (on-demand). Across 8 providers, the median rate card — each provider’s cheapest on-demand rate, community capacity and marketplace floors out — is $3.87 per GPU-hour, and the middle half of those rate cards sits between $3.59 and $4.87. Between rate cards, on-demand rates for the same card run from $3.32 (IONOS Cloud) to $5.50 (EmpirioLabs), 1.7× apart across 8 providers. By market, the cheapest listed rates are $3.94 on community capacity (Vast.ai, floor) and $3.32 on-demand (IONOS Cloud). With 141 GB of VRAM, that is $0.024 per GB of memory per hour at the cheapest on-demand rate and $0.027 at the median.
As of , one provider prices the NVIDIA H200 NVL on-demand as a whole 8-GPU node: $43.74 per node-hour (Tata Communications, BareMetal NVIDIA H200 NVL, 8X, $5.47 per GPU). 8 GPUs at the cheapest on-demand rate ($3.32 on IONOS Cloud) come to $26.52 an hour, where a provider rents 8 to one account and they need not share a unit. A unit price is what the provider quotes for the 8 GPUs together; whether it also rents one is not part of the quote.
8 GPUs for a month, 24/7 (730 h) cost from $19,363 at the listed on-demand rate of IONOS Cloud ($3.32 per GPU-hour); the job bar above ranks every provider for a job of your own.
Renting the H200 NVL for a year or more?
No provider we read publishes a committed H200 NVL rate: reserved H200 NVL capacity is sold through sales teams, at prices that are not public. A request goes to the providers with the cheapest listed on-demand H200 NVL rate.
Across every card, a 1-year rate runs a median −16% against the same provider’s on-demand rate (6 providers); a 3-year rate −41% (4 providers). Every committed rate, card by card.
Get quotes for the H200 NVL. A person at GPU Economy reads your request and passes it to up to three providers that list a rate for the card, the cheapest listed first; the quotes come from them, directly to you.
We may be paid a fee if a request becomes a contract; who receives it follows the listed rates, never a fee, and it changes no published rate or the Price Index (how, privacy).
Weekly history since 2025, with a like-for-like table per provider: NVIDIA H200 NVL price history.
Retail listings
No retail listings tracked for this card in the last 30 days.
Every rate on this page as data: /gpu/h200-nvl/prices.json — the same fields as prices.json, narrowed to this card. CC BY 4.0, no key, no rate limit. On your own page: embed this card’s cheapest rates (/embed/gpu/h200-nvl, an iframe, no script).
Data updated: