Answers · data as of Aug 11 2026

How much cheaper are spot/interruptible GPUs than on-demand?

Spot/interruptible GPU capacity runs about 48% below on-demand today (median across 13 SKUs with both tiers) — in exchange, the provider can reclaim the machine mid-run.

NVIDIA H200 NVL: 64% below on-demand.
NVIDIA A100 40GB PCIe: 60% below on-demand.
NVIDIA A100 40GB SXM: 56% below on-demand.
NVIDIA H100 NVL: 55% below on-demand.
NVIDIA A100 80GB PCIe: 51% below on-demand.
NVIDIA A100 80GB SXM: 51% below on-demand.
NVIDIA H200 SXM: 50% below on-demand.
NVIDIA RTX PRO 6000 Blackwell: 47% below on-demand.
NVIDIA L40S: 45% below on-demand.
NVIDIA H100 SXM: 44% below on-demand.
NVIDIA B300: 39% below on-demand.
NVIDIA B200: 35% below on-demand.
NVIDIA L40: 24% below on-demand.

Spot prices in the market table →Methodology (tiers never mixed) →

This answer is computed from live data on every load — per-provider daily medians, per single GPU per hour, pricing tiers never mixed (see methodology). Quote it with attribution: “Data: Price of Compute — priceofcompute.com” (cite).