GPU compute prices
Glossary
Answer in brief
Terms
- On-demand
An on-demand price is what you pay to start a machine when you want it and stop it when you are done, with no term and no risk of losing it.
- Spot
A spot price is a lower price for spare capacity that the provider can take back at short notice.
- Reserved
A reserved price is a lower hourly rate in return for a commitment to a term, often paid ahead, and billed for every hour of that term.
- Node
A node is one physical machine with several GPUs in it, most often eight, rented as a whole.
- Quota
A quota is the number of machines of a given kind that your account is allowed to run, and for large GPUs a new account often starts at zero.
- Per-second billing
Per-second billing charges for the exact time a machine ran, to the second, at the hourly rate divided down.
- GPU-hour
A GPU-hour is one GPU running for one hour, and it is the unit that lets machines of different sizes be compared on one line.
- Egress
Egress is data leaving a provider’s network, which some providers bill by the gigabyte and others include in the price.
Questions and answers
Sources for this entry
- Runcrate: Runcrate’s official site: what the platform is, how it bills and the range of GPUs it offers. Read 12 Nov 2026.
- Runcrate pricing: Runcrate’s live price feed: each GPU it lists, whether one can be started now, and the price it starts from per GPU-hour, read when a page is built. Read 12 Nov 2026.
- Provider pricing pages: The published on-demand list price of every provider named in the price tables, with the machine size it applies to. Read 12 Nov 2026.
- Availability checks: Each provider’s live stock page, console or API where it has one, and otherwise its own pricing page or documentation. Read 12 Nov 2026.
Checklists are original editorial guidance. Published prices do not establish availability.
Continue this reading path
Cite this entry
Compute Price Watch. “Glossary.” Content reviewed 12 Nov 2026. https://computepricewatch.com/glossary