RRunyard Index
Runyard / GPU Index · LLM API Index →

GPU Rental Price Index

Last collected 2026-10-11 · 84 snapshots since 2025-06-04

H100 SXM by provider type
Median by provider type, per snapshot
HyperscalersNeocloudsLambda (archive, pre-live)Marketplaces
$0.00$2.50$5.00$7.50$10.00Oct 5Oct 7Oct 9Oct 11

Median price by provider type, per snapshot. Hover for values. Lambda's archive history reaches back to June 2025 where it sold the GPU; everything else starts 2026-10-11.

H100 SXM median
$3.50/hr
B200 median
$6.99/hr
Providers tracked
49
GPUs priced today
32of 32

Datacenter GPUs use HBM memory and are sold in multi-GPU nodes. They carry the CME futures and set the market.

1NVIDIANVIDIA H100 SXM80 GB HBM3 · Hopper · DatacenterThe reference datacenter GPU. Underlies the CME H100 rental index.Median$3.50/ GPU / hr+39 providers2NVIDIANVIDIA A100 80GB80 GB HBM2e · Ampere · DatacenterThe previous-generation workhorse, still widely rented and steadily cheaper.Median$1.79/ GPU / hr+31 providers3NVIDIANVIDIA H200141 GB HBM3e · Hopper · DatacenterHopper with 141 GB of faster memory. Same compute as the H100, far more room for KV cache.Median$4.50/ GPU / hr+27 providers4NVIDIANVIDIA L40S48 GB GDDR6 · Ada Lovelace · DatacenterAda datacenter card for inference and graphics. 48 GB without HBM pricing.Median$1.29/ GPU / hr+23 providers5NVIDIANVIDIA B200180 GB HBM3e · Blackwell · DatacenterThe Blackwell flagship for training and large-scale inference. Underlies the CME B200 rental index.Median$6.99/ GPU / hr+21 providers6NVIDIANVIDIA L424 GB GDDR6 · Ada Lovelace · DatacenterLow-power inference card. 24 GB, single slot.Median$0.79/ GPU / hr+14 providers7NVIDIANVIDIA B300288 GB HBM3e · Blackwell Ultra · DatacenterBlackwell Ultra. The largest single-GPU memory NVIDIA ships, built for serving the biggest models.Median$8.00/ GPU / hr+13 providers8NVIDIANVIDIA A100 40GB40 GB HBM2 · Ampere · DatacenterThe original A100.Median$1.82/ GPU / hr+11 providers9NVIDIANVIDIA V100 16GB16 GB HBM2 · Volta · DatacenterThe original V100.Median$0.78/ GPU / hr+9 providers10NVIDIANVIDIA L4048 GB GDDR6 · Ada Lovelace · DatacenterAda datacenter card, graphics-leaning sibling of the L40S.Median$0.88/ GPU / hr+6 providers11NVIDIANVIDIA H100 PCIe80 GB HBM2e · Hopper · DatacenterThe PCIe H100. Same memory as the SXM, roughly 60% of its bandwidth.Median$2.29/ GPU / hr+4 providers12NVIDIANVIDIA T416 GB GDDR6 · Turing · DatacenterThe cheapest datacenter GPU still widely available.Median$0.59/ GPU / hr+4 providers13NVIDIANVIDIA A1024 GB GDDR6 · Ampere · DatacenterAmpere inference and graphics card.Median$1.29/ GPU / hr+3 providers14NVIDIANVIDIA A4048 GB GDDR6 · Ampere · DatacenterAmpere 48 GB datacenter card, now a budget option.Median$0.59/ GPU / hr+3 providers15AMDAMD MI300X192 GB HBM3 · CDNA 3 · Datacenter192 GB per GPU, more than any H100 or H200. The capacity play.Median$3.02/ GPU / hr+2 providers16NVIDIANVIDIA H100 NVL94 GB HBM3 · Hopper · DatacenterPCIe H100 with 94 GB, sold in NVLink pairs.Median$2.80/ GPU / hr+2 providers17NVIDIANVIDIA H200 NVL141 GB HBM3e · Hopper · DatacenterPCIe H200 for air-cooled servers.Median$3.79/ GPU / hr+1 providers18NVIDIANVIDIA GH20096 GB HBM3 · Grace Hopper · DatacenterHopper GPU fused with a Grace CPU and shared memory.Median$3.45/ GPU / hr4 providers19NVIDIANVIDIA V100 32GB32 GB HBM2 · Volta · DatacenterVolta, two generations back, still rented at the low end.Median$2.90/ GPU / hr4 providers20NVIDIANVIDIA A10G24 GB GDDR6 · Ampere · DatacenterThe AWS-specific A10 variant found in g5 instances.Median$1.52/ GPU / hr2 providers21AMDAMD MI355X288 GB HBM3e · CDNA 4 · DatacenterAMD's Blackwell-generation answer. 288 GB.Median$8.60/ GPU / hr1 provider22AMDAMD MI325X256 GB HBM3e · CDNA 3 · DatacenterMI300X with more, faster memory.Median$3.80/ GPU / hr1 provider

Median = median of each provider's own on-demand median, so a provider with thirty instance sizes counts once. Logos are the cheapest providers for that GPU. Ranked by provider coverage, then price — never by our traffic.

More from the index

ProvidersEvery provider against the market, with history and price changesCompareH100 vs A100, H200 vs B200, RunPod vs Lambda…What compute costsOne table: median, cheapest, spot, per month, per million tokensBy countryUSA, UK, India, Germany, France… each in its own currency

Providers

ProviderTypeHow prices are collected
Amazon Web ServiceshyperscalerPublic on-demand price list for Linux, us-east-1. Instance price divided by GPU count.
Microsoft AzurehyperscalerAzure Retail Prices API, Linux, eastus. Pay-as-you-go and Spot recorded separately. VM price divided by GPU count.
RunPod Secure CloudneocloudPublic GPU-type listing. Secure Cloud is RunPod's own datacentre capacity, quoted per GPU-hour.
RunPod Community CloudmarketplaceThird-party hosts renting through RunPod. Quoted per GPU-hour; a listing may have no hosts behind it.
LambdaneocloudPublic on-demand pricing page. Instance price divided by GPU count.
Vast.aimarketplacePublic offer search, verified and rentable hosts only. Each offer is one observation; on-demand rate, not interruptible bids.
Oracle CloudhyperscalerOCI's public price-list API (the cost estimator's feed), pay-as-you-go, per GPU-hour as listed. One price worldwide.
CoreWeaveneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
Massed ComputeneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
CivoneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
KoyebneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
OblivusneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
DataCrunch (Verda)neocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
GPU.aimarketplacePublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
ReplicateneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
Paperspace (DigitalOcean)neocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
BeamneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
Latitude.shneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
SpheronmarketplacePublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
RuncrateneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
LyceumneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
fal.aineocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
EnvergeneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
HyperAIneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
Packet.aineocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
Cyfuture AIneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
VultrhyperscalerPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
CrusoeneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
HyperstackneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
NebiusneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
Together AIneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
ModalneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
JarvislabsneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
Thunder ComputeneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
GMI CloudneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
BasetenneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
DigitalOcean GPU DropletshyperscalerPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
Lightning AIneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
SaladCloudmarketplacePublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
ScalewayhyperscalerPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
CerebriumneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
Taiga CloudneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
NeysaneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
E2E NetworksneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
TensorDockmarketplacePublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
io.netmarketplacePublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
AkashmarketplacePublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
NorthflankneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
QubridneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
SeewebneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
OVHcloudhyperscalerPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.
Denvr DataworksneocloudPublic pricing page, parsed. Per-GPU price is the listed per-GPU rate, or the node price divided by GPU count. Prices near reserved, committed or contract terms are discarded; anything outside a plausible range for the GPU is discarded.

Questions

How much does it cost to rent a GPU right now?

As of 2026-10-11, across 49 providers: the H100 SXM has a market median of $3.50 per GPU-hour (cheapest listing $1.68 at Latitude.sh), the B200 $6.99, and the cheapest GPU-hour of any kind is the NVIDIA RTX A4000 at $0.16. Multiply by 720 for a month. Every GPU in the list has its own page with every provider's price.

What is the "median" and why not the cheapest?

The median is the middle of the market: each provider is first reduced to the median of its own prices for that GPU, so a provider with thirty instance sizes counts once, then the median is taken across providers. The cheapest listing is shown too, but it is usually a single marketplace host and can be gone in an hour. The median tells you whether a quote is high or low; the cheapest tells you the floor.

Where do the prices come from?

Every provider's own public pricing — their API where one exists, their pricing page where not — read every six hours. 49 providers, 1106 priced configurations today. No aggregator's data is licensed or scraped; the methodology page lists every source and the full calculation.

Why is the same GPU $2 at one provider and $11 at another?

Three markets share one name. Hyperscalers price the instance — CPU, RAM, network and the ecosystem — so the GPU is a fraction of it. Neoclouds price the GPU. Marketplaces are independent hosts competing on price with consumer-grade guarantees. The chart at the top stacks the three so the premium is visible rather than hidden.

How often does it update, and is there history?

Every six hours, and every snapshot is kept — 84 so far, since 2025-06-04. Each GPU page charts its median over time by billing type, each provider page charts its own prices, and price changes between snapshots are logged. Nothing is interpolated: a gap in collection is a gap on the chart.

Can I see prices in my currency?

Yes — the currency picker in the top bar converts every figure at the rate captured with the latest snapshot, and each country page prices the whole market in that country's currency against the global median.

Want to know which of these can actually hold a specific model? Every GPU page shows what fits at 4, 8 and 16-bit — and what the set costs per hour.
Start with the H100 →