← GPU price tracker

MI325X GPU Rental Price

The AMD MI325X (CDNA 3, 256 GB HBM3e) rents from $2.00/hr, averaged across 3 cloud providers, updated 2026-08-14.

$3.18
on-demand /hr
$2.00
spot /hr
$2.03
reserved /hr
$2,321
/mo on-demand

Memory-maxed CDNA 3 part (256 GB) aimed at large-model inference where capacity, not raw FLOPs, is the bottleneck. CDNA 3 · 256 GB HBM3e · 6 TB/s memory bandwidth · launched 2025.

on-demand /hr · spot /hr

Recent daily prices

DateOn-demandSpot
2026-08-14$3.18$2.00
2026-08-13$3.18$2.00
2026-08-12$3.18$2.00
2026-08-11$3.18$2.00
2026-08-10$3.18$2.00
2026-08-09$3.18$2.00
2026-08-08$3.18$2.00
2026-08-07$3.18$2.00
2026-08-05$3.18$2.00
2026-08-04$3.18$2.00
2026-08-03$3.18$2.00
2026-08-02$3.18$2.00
2026-08-01$3.06$2.00
2026-07-31$3.06$2.00

MI325X vs other GPUs

GPUOn-demand /hrSpot /hrMemory
MI325X$3.18$2.00256 GB HBM3e
B200$6.84$3.74192 GB HBM3eMI325X vs B200 →
H200$4.37$2.98141 GB HBM3eMI325X vs H200 →
H100$4.00$1.7180 GB HBM3MI325X vs H100 →
MI355X$8.60$5.29288 GB HBM3eMI325X vs MI355X →
MI300X$3.56$1.65192 GB HBM3MI325X vs MI300X →

FAQ

How much does it cost to rent an MI325X per hour?

The MI325X averages $3.18/hr on-demand, $2.00/hr spot, and $2.03/hr reserved across 3 cloud providers (updated 2026-08-14).

What is the cheapest MI325X cloud price?

The lowest tier is spot at about $2.00/hr on average; reserved (1–3 year commitment) runs ~$2.03/hr. Compare live per-provider rates on the Bargo GPU price tracker.

How much is an MI325X per month?

At on-demand $3.18/hr, one MI325X runs about $2,321/month (730 hours); reserved pricing lowers that to roughly $1,482/month.

Is the MI325X cheaper than the H100?

On-demand, the MI325X is $3.18/hr vs the H100 at $4.00/hr — cheaper per hour. See the full MI325X vs H100 comparison for specs and cost-per-job context.

What's the difference between MI325X spot and on-demand pricing?

Spot ($2.00/hr) is interruptible spare capacity at a discount to on-demand ($3.18/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.