Skip to content
InfratoApp
GPU prices / H200-SXM-141GB

NVIDIA H200 SXM 141 GB

H200-SXM-141GB costs $2.09 to $6.30 per GPU-hour on demand across 26 cloud providers, cheapest at Beam. Prices were read from each provider's own pricing page, most recently on 2026-10-27.
Cheapest read 27 Oct · spread 3.0× · a missing provider is never an estimate
CHEAPEST $2.09/hrMEDIAN $4.33/hrSPREAD 3.0×PROVIDERS 26

Price by provider — cheapest first

Provider$/GPU·hr ↓Δ 7dTermsRead
Beam$2.09— 0.0%not stated27 Oct ↗
Shadeform$2.25— 0.0%not stated26 Oct ↗
Theta EdgeCloud$2.29— 0.0%not stated26 Oct ↗
Sesterce$2.48FIRST READINGnot stated7 Oct ↗
Koyeb$3.00— 0.0%not stated26 Oct ↗
Civo$3.49— 0.0%on demand27 Oct ↗
DeepInfra$3.49— 0.0%not stated26 Oct ↗
Lium$3.90▼ −17.0%not stated27 Oct ↗JSON
Hyperbolic$3.99FIRST READINGnot stated26 Oct ↗
Oblivus$3.99— 0.0%on demand26 Oct ↗
Hyperstack$3.99— 0.0%not stated26 Oct ↗
Canopy Wave$4.00○not stated27 Oct ↗
Cerebrium$4.20— 0.0%not stated27 Oct ↗
DigitalOcean$4.47FIRST READINGon demand11 Oct ↗
Modal$4.54— 0.0%not stated26 Oct ↗
Vast.ai$4.60○on demand27 Oct ↗
Runcrate$4.63▲ +9.2%not stated26 Oct ↗
Spheron$4.80▼ −12.9%not stated26 Oct ↗
RunPod$5.29— 0.0%on demand27 Oct ↗
Replicate$5.49— 0.0%not stated26 Oct ↗
Enverge$5.50— 0.0%not stated26 Oct ↗
Crusoe$6.00▼ −20.0%not stated27 Oct ↗
Fal.ai$6.00— 0.0%not stated26 Oct ↗
Verda (DataCrunch)$6.07▲ +5.1%on demand27 Oct ↗
Lyceum$6.19— 0.0%on demand26 Oct ↗
CoreWeave$6.30— 0.0%on demand26 Oct ↗
Why a provider might be missing: its price is behind a quote form, older than 21 days, or flagged for review. A missing row is never an estimate.

Price history — cheapest daily reading

⚑ 7 of 12 weeks measuredWEEKLY LOW SHOWN
2026-08-04A FLAT MARK IS A WEEK WE DID NOT READ2026-10-20

Cost calculator

GPUs1×
Hours / month730 h
730 = 24/7 · on-demand rates only
Beam$1,526 /mo
Shadeform$1,643 /mo
Theta EdgeCloud$1,672 /mo
Sesterce$1,810 /mo
Koyeb$2,190 /mo
Civo$2,548 /mo
DeepInfra$2,548 /mo
Lium$2,847 /mo
Hyperbolic$2,913 /mo
Oblivus$2,913 /mo
Hyperstack$2,913 /mo
Canopy Wave$2,920 /mo
Cerebrium$3,066 /mo
DigitalOcean$3,263 /mo
Modal$3,314 /mo
Vast.ai$3,358 /mo
Runcrate$3,380 /mo
Spheron$3,504 /mo
RunPod$3,862 /mo
Replicate$4,008 /mo
Enverge$4,015 /mo
Crusoe$4,380 /mo
Fal.ai$4,380 /mo
Verda (DataCrunch)$4,427 /mo
Lyceum$4,519 /mo
CoreWeave$4,603 /mo

Spec sheet

VRAM141 GB
Bandwidth4.8 TB/ssource
FP16989 TFLOPSsource
FP81979 TFLOPSsource
TDP700 Wsource
NVLinkNVLink 4source
ArchitectureHopper
Released2024source
8 of 8 figures held

Good for

Memory-bound 70-100B inference where extra HBM/bandwidth matter; no FP4 needed
reasoned from the specification, not measured · source

Benchmarks

dlrm-v2-99 Offline
643,122 Samples/s
v5.1 · 8 accelerators · ASUSTeK
submission ↗
dlrm-v2-99 Server
591,162 Queries/s
v5.1 · 8 accelerators · ASUSTeK
submission ↗
dlrm-v2-99.9 Offline
388,058 Samples/s
v5.1 · 8 accelerators · ASUSTeK
submission ↗
dlrm-v2-99.9 Server
370,066 Queries/s
v5.1 · 8 accelerators · ASUSTeK
submission ↗
llama3.1-8b Offline
163,292 Tokens/s
v6.1 · 16 accelerators · Cisco
submission ↗
llama3.1-8b Server
161,110 Tokens/s
v6.1 · 16 accelerators · Cisco
submission ↗

LLMs that fit this GPU’s VRAM

FP16
~56B params
FP8
~112B params
FP4/INT4
~225B params
Llama 3.1 70B @FP8; Qwen 72B @FP8; 32B @FP16
8-GPU = 1128GB pooled; fits 405B @FP8
ceilings at each precision, reasoned from VRAM · not a benchmark

Frequently asked

How much does an H200-SXM-141GB cost per hour right now?
$2.09 an hour at Beam, the cheapest of 26 sellers we read. The dearest posts $6.30 at CoreWeave. Every rate links to the page it was read from, with the date it was read.
Why do H200-SXM-141GB prices vary 3.0× between providers?
Because a GPU-hour is not one product. $2.09 at Beam and $6.30 at CoreWeave buy different things: interconnect, host CPU and RAM, storage, egress, support, and whether the capacity is dedicated or shared. We publish the posted rate for each and the page it came from, and do not blend them into one number.
What is the cheapest way to rent an H200-SXM-141GB long-term?
We hold no committed or reserved rate for this card — every figure here is an on-demand posted rate. Committed pricing is usually quoted rather than published, so the honest answer is that we cannot tell you, rather than a guess at a discount.
How fresh are these prices?
These rates were read between 2026-10-07 and 2026-10-27. Nothing older than 21 days is shown: a reading that goes stale is hidden rather than displayed with an old date, because a wrong price is worse than a missing one. Each row links to the page its figure came from, so any of them can be checked against the seller.
Which is cheaper for an H200-SXM-141GB — a region or a commitment?
Both move the number and they move it independently. Where a provider prices the same card in several regions we show the cheapest they charge anywhere and name the region, because a rate you can only get in Estonia is still a rate you can get. A committed term is a different contract rather than a discount on this one, so committed rates are kept out of the headline and shown separately.
Can a provider pay to be listed first for the H200-SXM-141GB?
No. The table has one stated sort at a time and nothing in it can be bought — a sponsor cannot enter it, reorder a row or change a rate. How the site is paid for, and the line each source of money does not cross, is set out under money and independence at /our-data#sec5.
Vs its class
Live cheapest rates
GPUCheapest $/hrVRAMFP8 TFLOPS
H200-SXM-141GB ← this$2.09141 GB○
H200-NVL-141GB$3.62141 GB○

Cite this page

infratoapp, “H200-SXM-141GB cloud pricing”, read 27 October 2026. $2.09 per GPU-hour at Beam. https://www.infratoapp.com/gpu/H200-SXM-141GB
Reading date travels inside the citation. How we read this data →