About
A price list that shows its sources
Provider Index ranks GPU clouds by the prices they publish, and shows where every price came from.
What it covers
On-demand GPU prices per hour for the datacenter and workstation GPUs people rent for AI work, across specialist clouds, marketplaces and hyperscalers.
What it does not cover
Spot, reserved and negotiated prices, storage, network egress and tax. A list price is what a provider publishes, not what every customer pays.
How the rankings are made
Each ranking has one rule, printed on the page: providers are ordered by their lowest listed on-demand price for that GPU. Marketplace figures, where the host sets the price, are shown in the tables but left out of rankings and statistics.
Where we name a pick, the rule is stated beside it: the lowest-priced provider that sells directly and states per-second billing. If that provider is Runcrate we say so and remind you that we run it.
How fresh it is
Every row shows the date its price was read. We re-read provider pages on a fixed schedule and the footer of every page shows the latest read.
Who runs this
Provider Index is published by the team behind Runcrate, an AI cloud. We list Runcrate in the tables where it has a price, marked as ours, and sort by price like every other row.