← GPU price tracker

B200 GPU Rental Price

The NVIDIA B200 (Blackwell, 192 GB HBM3e) rents from $4.80/hr, averaged across 45 cloud providers, updated 2026-11-15. On-demand pricing has risen 28% over the past 90 days.

$8.28
on-demand /hr
$4.80
spot /hr
$6.02
reserved /hr
$6,044
/mo on-demand
+24%
YoY

Blackwell-generation flagship — roughly 2–3× H100 training throughput at FP8/FP4 and a far lower cost-per-token on large models. Blackwell · 192 GB HBM3e · 8 TB/s memory bandwidth · launched 2025.

▮ on-demand /hr · ▮ spot /hr

Recent daily prices

DateOn-demandSpot
2026-11-15$8.28$4.80
2026-11-14$8.02$4.78
2026-11-13$8.18$4.78
2026-11-12$7.99$4.74
2026-11-11$8.15$4.66
2026-11-10$8.13$4.26
2026-11-09$8.25$4.26
2026-11-08$8.13$4.18
2026-11-07$8.13$4.42
2026-11-06$8.19$4.66
2026-11-05$7.99$4.49
2026-11-04$7.83$4.26
2026-11-03$8.35$4.57
2026-11-02$7.95$4.39

B200 vs other GPUs

GPUOn-demand /hrSpot /hrMemory
B200$8.28$4.80192 GB HBM3e
H200$4.59$3.36141 GB HBM3eB200 vs H200 →
H100$4.01$2.4080 GB HBM3B200 vs H100 →
MI355X$8.25$3.56288 GB HBM3eB200 vs MI355X →
MI325X$3.11$—256 GB HBM3eB200 vs MI325X →
MI300X$2.86$1.84192 GB HBM3B200 vs MI300X →

FAQ

How much does it cost to rent an B200 per hour?

The B200 averages $8.28/hr on-demand, $4.80/hr spot, and $6.02/hr reserved across 45 cloud providers (updated 2026-11-15).

What is the cheapest B200 cloud price?

The lowest tier is spot at about $4.80/hr on average; reserved (1–3 year commitment) runs ~$6.02/hr. Compare live per-provider rates on the Bargo GPU price tracker.

How much is an B200 per month?

At on-demand $8.28/hr, one B200 runs about $6,044/month (730 hours); reserved pricing lowers that to roughly $4,395/month.

Is the B200 cheaper than the H100?

On-demand, the B200 is $8.28/hr vs the H100 at $4.01/hr — pricier per hour. See the full B200 vs H100 comparison for specs and cost-per-job context.

What's the difference between B200 spot and on-demand pricing?

Spot ($4.80/hr) is interruptible spare capacity at a discount to on-demand ($8.28/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.