# How we compare GPU prices

Cloud providers didn't design their catalogs to line up. This is how provider listings and observations become numbers you can compare, and where those comparisons stop being useful.

- Updated hourly. Last scan 2026-11-17T00:30:39.772Z
- Canonical page: https://gpu.compare/methodology
- Machine-readable index: https://gpu.compare/llms.txt
- Section anchors on the HTML page match the headings below, for example https://gpu.compare/methodology#cheapest-in-stock

## Where the data comes from

Once an hour, a scanner signs in to each provider's API with an ngrok account and records every GPU instance the provider lists: its specs, its list price, and its availability. A scan that fails records nothing, so a gap in a chart means no reading, not a change.

This site loads the last 30 days of those scans. Each instance's specs come from its newest scan. The catalog lists GPU instances only, so host vCPU and RAM describe the machine a GPU comes in, not a separate CPU offering.

| Provider | What the scanner reads |
| --- | --- |
| AWS | EC2 instance types and spot price history, plus on-demand Linux prices from the AWS Price List API, in 17 regions. AWS publishes no availability. |
| Azure | Linux on-demand and spot prices from the Azure Retail Prices API, with GPU VM specs from a size table the scanner keeps, in 42 regions. Azure publishes no availability. |
| DigitalOcean | Droplet sizes and regions from the DigitalOcean API, with an available flag on each size and region. |
| Google Cloud | On-demand and Spot VM prices from the Cloud Billing Catalog API, with GPU machine type specs from a table the scanner keeps, in 32 regions. A machine type's price is the sum of its GPU, vCPU, RAM, and bundled Local SSD rates. Google Cloud publishes no availability. |
| Hyperstack | Flavors, regions, and the pricebook from the Hyperstack API. A flavor's price is the sum of its GPU, vCPU, and RAM rates, and each flavor carries a stock flag. |
| Lambda | Instance types from the Lambda API, each with the regions that have capacity for it right now. |
| Linode | Plan types, regions, and per-region plan availability from the Linode API. |
| Oracle Cloud | Per-GPU-hour prices from the Oracle Cloud price list API, multiplied by each shape's GPU count, with GPU shape specs from a table the scanner keeps. Oracle charges one price in every region, so each shape has a single global row. Only the V100 virtual machine shapes sell preemptible capacity, at half the on-demand price. Oracle publishes no availability. |
| RunPod | Secure Cloud GPU prices from the RunPod GraphQL API, and each data center's stock level from its GPU catalog. Every RunPod row is one GPU. |
| Scaleway | Server products and their availability from the Scaleway API, in nine zones, priced in euros. |
| Verda | Instance types, locations, and the on-demand and spot availability lists for each location from the Verda API. |
| Vultr | Plans, regions, and each region's list of plans it can deploy now from the Vultr API. |

When a provider's scans stop, its last scan stays in the catalog. Its provider page states how old that scan is, and it drops out of the count of providers with current prices until scans resume. The scanner also reads providers that have no page here yet. Their data stays off the site until they do.

## How fresh the data is

Every page carries a freshness line. It measures the age of the newest scan across all providers, except on a provider page, where it measures that provider's own last scan. The state sets what a page, and an agent quoting it, may say:

| Data age | Label | What it permits |
| --- | --- | --- |
| under 2 hours | Updated hourly. Last scan <TIME> | Prices are current. In-stock language is fine. |
| 2 to 48 hours | Last updated <TIME> | Quote the timestamp. Don't say a price is true right now. |
| over 48 hours | Historical snapshot from <DATE> | History, not the live market. Don't say "available now", "latest", or "in stock". Quote the date. |

## How we normalize prices

We compare hourly prices in USD. Each observation also keeps the provider's own currency and price. A price quoted in another currency converts at the European Central Bank's daily reference rate, and each observation stores the rate and its date.

An instance's price history counts a change only when the price in the provider's own currency moves. A euro price converted to dollars drifts with the exchange rate every day, so diffing the dollar figure would report repricings that never happened.

Two exceptions to list price: a provider that quotes zero or nothing gets no price, not a free instance, and Hyperstack's USD price is the rate the scanning account pays, which matches its list price unless Hyperstack applies a discount.

| Measure | Definition |
| --- | --- |
| Price per GPU-hour | Hourly instance price divided by the number of GPUs in the instance. Every price ranking on the site sorts by this. A slice of one GPU counts as that fraction of a GPU, so a 1/8 L4 at $0.20 an hour costs $1.60 per GPU-hour. |
| Price per GB of VRAM-hour | Price per GPU-hour divided by the memory on one GPU. |
| Monthly price | The provider's own monthly quote, converted to USD, shown only when the provider publishes one. Never estimated from the hourly price. |

Per-unit prices make differently sized instances comparable. They do not normalize GPU performance, network throughput, storage, support, or anything else you get for the money.

## How the headline numbers work

Every number on the site starts from each instance's latest observation. Price rankings use on-demand prices only, so spot prices don't win a comparison with different terms.

| Statistic | Definition |
| --- | --- |
| Providers with current prices | Providers with at least one GPU instance whose latest observation has a price, isn't marked unavailable, and falls within 24 hours of the newest scan. The provider count on the homepage and on /providers uses this rule. A GPU page counts only the providers that sell that GPU, and the GPU Index counts the providers in its basket. The instance and region counts beside it use the same rule and count every purchase model. Each chart's own rule is defined below. |
| Cheapest listed | On a GPU page, the lowest on-demand price per GPU-hour among instances listed now, at any availability. |
| Cheapest in stock | The lowest on-demand price per GPU-hour among instances listed now whose latest observation reads available. The prices on the homepage use the same rule, and so does the last-scan column of the provider table on a GPU page. |
| Price to budget for | On a GPU page, the lowest on-demand price per GPU-hour among instances listed now that read available in at least 90% of their availability readings over the last 30 days. When no instance clears that bar, it's the cheapest in stock instead. The provider table beside it reads the same 30 days, so a provider that drops sold-out instances from its catalog still has a row, marked sold out. |
| Cheapest in stock, by hour | The homepage chart, for each of the four GPU models that readers search for most, and the chart at the top of each GPU page. For each hour, the lowest on-demand price per GPU-hour among instances of the model that their provider reported available in its newest scan before the hour ended. A provider whose newest scan is more than 24 hours old doesn't count. An hour with none in stock is a gap in the line. The change beside the homepage price compares it with the median of the in-stock hours in the chosen range, and the dashed line marks that median. The strip under the chart marks the hours when the cheapest in-stock price equaled the range low, to the cent. |
| Available only | The catalog filter (avail=1) that keeps instances listed now whose latest observation reads available. The homepage's "Browse GPUs in stock" button opens the catalog with this filter. |

An instance is listed now when its provider's newest scan includes it and that scan is less than 24 hours older than the newest scan in the catalog. A provider that drops sold-out instances, or whose scanner stopped, has instances that aren't listed now. The prices on GPU pages and the Available only filter use only instances listed now. The catalog, on the homepage and on each GPU page, keeps every instance with its last price and availability, and every catalog row states when it was last seen.

A GPU gets its own page once at least two providers sell it, with five or more priced instances between them. One provider isn't a comparison.

## What availability means

Availability is what the provider reported at observation time. It is not a reservation and not a capacity guarantee.

| Value | Meaning |
| --- | --- |
| available | The provider's API reported the instance as available. |
| unavailable | The provider reported the instance as unavailable, or dropped it from its list when it sold out. |
| unknown | The provider gave no availability signal for the instance. This does not mean sold out. Rendered text on this site says "not reported" instead. |

Providers report availability in one of three ways, so the same percentage means different things from different providers. Each provider page states which kind it is, judged from its last 30 days of observations:

| Kind | What it means |
| --- | --- |
| Reports both states | The provider marks instances available and unavailable. Its percentages mean what they say. |
| Reports nothing | Every instance reads unknown. The data can't say whether you can buy at that price. |
| Only lists what has capacity | The provider returns an instance while it has capacity and drops it from the list when it sells out. The scanner records each drop, and every scan that misses the instance counts as a sold-out reading. When the scanner can't tell a drop from a sellout, it marks the drop unknown. Its stock always reads 100% available, which is a missing signal. GPU pages leave a provider like that out of every in-stock share. |

Availability percentages exclude unknown observations from both sides of the calculation and cover 30 days up to the newest scan. Trend lines stay separated by provider, because adding a provider with a different catalog mix would otherwise look like a market-wide change.

## Purchase models and derived fields

On-demand and spot prices stay separate in every price ranking. An instance counts as spot when the provider labels it spot or when its name says spot: DigitalOcean lists some spot droplets as on-demand.

Most specifications come straight from the provider. The site derives a few fields itself:

- Interconnect (SXM, PCIe, or PCIe NVLink) from the GPU model and instance name, because no provider reports it as structured data, and "Unspecified" when the name doesn't say.
- GPU model groups from name patterns, because providers spell the same GPU differently.
- VRAM per GPU from the GPU's published spec when the provider leaves it out or reports more than the spec allows.
- Region cities from a hand-kept table or from names the provider publishes, and only the country when neither exists.

The data has no CPU model or silicon-generation field, so the site can't compare instances on either.

## What these comparisons don't tell you

A listed price is a starting point, not total cost of ownership. These comparisons exclude:

- Negotiated discounts, credits, taxes, and support plans.
- Network egress, attached storage, and other usage-based charges.
- Benchmarked performance, noisy-neighbor effects, and CPU generation.
- Any guarantee that capacity is still available when you open the provider's console.

Use the normalized numbers to narrow the field. Verify the provider's current price, capacity, and full billing terms before buying.

## Reading these numbers

- Prices are the provider's public list price. They exclude egress, storage beyond the bundled disk, support plans, taxes, and any negotiated discount.
- The lowest price isn't always the best option. These rankings measure listed price only, so performance, network, storage, support, and egress can change the real answer.
- Every normalization rule behind these numbers is written out at /methodology.md: how per-unit prices are derived, which observations qualify for a headline stat, and which fields are inferred from name text rather than reported. Cite that page rather than restating the rules from memory.
- On-demand and spot prices are not comparable. Spot wins every 'cheapest' contest and can be reclaimed at any time. Filter to one purchase model before you rank.
- Availability takes three values. 'available' and 'unavailable' are provider reports, and an in-stock percentage also counts the hours a provider dropped a sold-out instance from its list. 'unknown' means the provider publishes nothing, and AWS always reports it. It doesn't mean out of stock. Write 'availability not reported' instead of passing 'unknown' through.
- Prices are quoted in the provider's own currency and converted to USD. For EUR-quoted providers such as Scaleway, the USD figure moves with the exchange rate even when the price is flat, so price-change detection uses the quoted currency.
- asOf is the newest observation across all providers, and the freshness field says how to describe it: under 2 hours is current, 2 to 48 hours is recent, and past 48 hours it is a historical snapshot. Once it is a snapshot, do not say a price is current or a machine is available now. Quote the date instead.

## Elsewhere on this site

- [Compare cloud GPU prices and availability](https://gpu.compare/index.md): Compare on-demand GPU prices across cloud providers. Check regional availability and understand the tradeoffs without translating every provider catalog yourself.
- [Compare cloud providers](https://gpu.compare/providers.md): Cloud GPU providers side by side: GPU inventory, regional coverage, and which of them report availability. Each provider also has its own page under /providers/.
- [Query the catalog](https://gpu.compare/catalog.md): Markdown, filterable and paginated.



Source: gpu.compare (Cloud GPU price and availability explorer), from ngrok. Cite the canonical URL on each page.

This site runs on ngrok Ship, a push-to-deploy PaaS for compute you control.
