Spot vs On-Demand GPU Pricing: A100 and H100 Rates Compared

Last reviewed on 2026-08-28 · 8 min read · Rates from the ComputingPower.org dataset and provider spot pages

The difference between spot and on-demand GPU pricing is the single largest price gap in cloud compute — larger than the gap between providers and larger than the gap between GPU generations. This page puts the two side by side for the same GPU at the same provider, adds the reserved and negotiated-contract tiers so you can see the full ladder, and works through when the spot discount is real money and when interruptions eat it.

−50% to −71%
Spot discount vs on-demand, A100
$0.51
Lowest A100 spot / hr in dataset
$2.29
Lowest H100 on-demand / hr
10 s – 5 min
Termination notice range

What the two prices are

On-demand is the public list price: a guaranteed instance, billed per second or per hour, that you can keep for as long as you pay. Spot (Google calls it Spot VM, formerly preemptible; marketplaces call it "interruptible") is the same hardware sold from the provider's unused capacity. Yes — spot rates are low precisely because the capacity would otherwise sit idle earning nothing, and the provider reserves the right to reclaim it for a full-price customer with seconds to minutes of notice. You are paying less in exchange for accepting that risk; the hardware and performance are identical.

Spot vs on-demand, same GPU, same provider

Illustrative on-demand and spot rates per GPU-hour. Spot prices float with demand; the figures are representative of the range seen during review, not a live quote.

ProviderGPUOn-demandSpotSavingTermination noticeTypical spot availability
AWS EC2A100 40GB (p4d, per GPU)$3.22$0.97−70%2 minutesModerate — pools drain fast in us-east-1
Google CloudA100 40GB (a2)$2.77$0.83−70%30 secondsGood in secondary regions
AzureA100 80GB (ND A100 v4, per GPU)$3.67$1.47−60%30 secondsModerate
Lambda LabsA100 40GB$2.40$1.20−50%5 minutesGood
Vast.aiA100 40GB$1.79$0.51−71%None guaranteed (outbid = gone)Variable by host
RunPodA100 80GB$1.89$0.76−60%10 secondsGood

Three patterns are visible. Hyperscalers have the highest on-demand rate and the deepest percentage discount, so their spot price lands close to a specialist cloud's on-demand price. Marketplaces have the lowest absolute spot price but the weakest interruption guarantee. And the spot-to-on-demand ratio is remarkably stable at 30–40% of list, whichever provider you look at — that ratio is a better planning number than any single quote.

The full ladder: spot, on-demand, reserved and contract for the H100

Spot and on-demand are two rungs of a longer ladder. Using the H100 80GB, where the price index has the most listings:

TierH100 $/GPU-hrvs specialist on-demandCommitmentGuarantee
Hyperscaler on-demand$3.99 – 4.99+35% to +68%NoneFull
Specialist-cloud on-demand (index median)$2.97NoneFull
Marketplace on-demand (index low)$2.29−23%NoneFull, host-dependent
1-year reserved / committed use≈ $2.00 – 2.40−20% to −35%12 monthsFull
3-year reserved≈ $1.60 – 2.00−35% to −45%36 monthsFull
Spot / interruptible≈ $1.00 – 1.80−40% to −65%NoneNone — reclaimable
Negotiated multi-year cluster contractBelow public ratesNot publishedMulti-year, minimum spendFull, with priority on scarce SKUs

Two consequences. The spot-versus-contract spread is narrower than the spot-versus-on-demand spread: a three-year reservation gets most of the way to spot pricing with none of the interruption risk, which is why large inference fleets run reserved, not spot. And on a given day spot can briefly cost more than a reservation when a region is short of capacity — spot is a floating price, the other rungs are fixed.

When spot is actually cheaper

The headline discount is not what you save. Every interruption costs you the work since the last checkpoint plus the time to reprovision and reload, so the effective spot price is:

Effective spot $/hr = spot rate × (1 + wasted fraction), where wasted fraction ≈ (interruptions per hour) × (average minutes lost per interruption ÷ 60).

Worked example on an A100 at $0.97 spot versus $3.22 on-demand (the AWS row). If you checkpoint every 15 minutes, lose on average half an interval (7.5 min) plus 5 minutes to recover, and get interrupted once every four hours, the wasted fraction is 0.25 × 12.5/60 ≈ 5%. Effective spot cost is $1.02/hr — still 68% below on-demand. Push interruptions to once an hour with 30-minute checkpoints and the waste rises to ~33%; effective cost $1.29/hr, still a 60% saving. Spot loses only when either the job cannot checkpoint at all, or an SLA makes any interruption unacceptable.

The break-even rule of thumb: spot wins for any restartable job as long as (minutes lost per interruption × interruptions per hour) < 60 × (1 − spot/on-demand). At a typical 30–40% spot ratio, that leaves room for a great deal of interruption before spot stops paying.

Which price to use for which workload

WorkloadBest tierWhy
Pre-training, fine-tuning, hyperparameter sweepsSpot (with checkpoints)Restartable, throughput-bound, large GPU-hour count — the discount compounds
Batch inference, data preprocessing, evaluationSpotEmbarrassingly parallel; a lost worker costs one shard
Development, notebooks, short experimentsOn-demandInterruptions cost human time, which is worth more than the saving
Production inference endpointReserved / committed for the baseline, on-demand for peaksLatency SLA; spot variance is unacceptable
Hundreds of GPUs for a year or moreNegotiated contractGuaranteed access to scarce SKUs at below-public pricing

Related reading