← GPU price tracker

H200 GPU Rental Price

The NVIDIA H200 (Hopper, 141 GB HBM3e) rents from $3.36/hr, averaged across 54 cloud providers, updated 2026-11-15. On-demand pricing has risen 16% over the past 90 days.

$4.59
on-demand /hr
$3.36
spot /hr
$3.28
reserved /hr
$3,351
/mo on-demand
+12%
YoY

An H100 with ~1.8× the memory and bandwidth — same core compute, but fits larger models and serves long-context inference more efficiently. Hopper · 141 GB HBM3e · 4.8 TB/s memory bandwidth · launched 2024.

▮ on-demand /hr · ▮ spot /hr

Recent daily prices

DateOn-demandSpot
2026-11-15$4.59$3.36
2026-11-14$4.61$3.35
2026-11-13$4.60$3.39
2026-11-12$5.00$3.37
2026-11-11$5.00$3.42
2026-11-10$5.29$3.33
2026-11-09$4.93$3.41
2026-11-08$5.06$3.50
2026-11-07$5.00$3.46
2026-11-06$5.14$3.59
2026-11-05$5.13$3.33
2026-11-04$5.00$3.96
2026-11-03$4.67$3.16
2026-11-02$4.94$3.16

H200 vs other GPUs

GPUOn-demand /hrSpot /hrMemory
H200$4.59$3.36141 GB HBM3e
B200$8.28$4.80192 GB HBM3eH200 vs B200 →
H100$4.01$2.4080 GB HBM3H200 vs H100 →
MI355X$8.25$3.56288 GB HBM3eH200 vs MI355X →
MI325X$3.11$—256 GB HBM3eH200 vs MI325X →
MI300X$2.86$1.84192 GB HBM3H200 vs MI300X →

FAQ

How much does it cost to rent an H200 per hour?

The H200 averages $4.59/hr on-demand, $3.36/hr spot, and $3.28/hr reserved across 54 cloud providers (updated 2026-11-15).

What is the cheapest H200 cloud price?

The lowest tier is spot at about $3.36/hr on average; reserved (1–3 year commitment) runs ~$3.28/hr. Compare live per-provider rates on the Bargo GPU price tracker.

How much is an H200 per month?

At on-demand $4.59/hr, one H200 runs about $3,351/month (730 hours); reserved pricing lowers that to roughly $2,394/month.

Is the H200 cheaper than the H100?

On-demand, the H200 is $4.59/hr vs the H100 at $4.01/hr — pricier per hour. See the full H200 vs H100 comparison for specs and cost-per-job context.

What's the difference between H200 spot and on-demand pricing?

Spot ($3.36/hr) is interruptible spare capacity at a discount to on-demand ($4.59/hr); use spot for checkpointed or fault-tolerant jobs and on-demand or reserved for long, uninterrupted runs.

Cross-provider averages from public cloud GPU listings. Informational only.