RunPod · Service Levels

RunPod service levels

Figures are quoted from the pages listed in sources. Guarantees are what the provider promises. Observed figures are what the provider reports about itself. APIs.io runs no monitors and has not measured or verified any of them. Read 2026-11-02.

Promised uptime

Commitments on the provider's own SLA page.

Not checked: we hold no SLA page for this provider.

Reported uptime

What the provider says about itself, usually on its status page. It is not a promise, and periods differ, so don't compare or rank these.

100% 100% uptime

Serverless: api.runpod.ai · source

100% 100% uptime

Serverless: queue engine · source

100% 100% uptime

Serverless: CPU workers · source

99.999% 99.999% uptime

GPU Cloud: graphql: api.runpod.io · source

100% 100% uptime

GPU Cloud: log/metrics api: hapi.runpod.net · source

100% 100% uptime

GPU Cloud: pod proxy: proxy.runpod.net · source

100% 100% uptime

GPU Cloud: graphql: transactional data store · source

100% 100% uptime

GPU Cloud: ui: runpod.io/console · source

Document

Download YAML · format: SLA4OAI 1.0.0 Draft, extended with x-apis-io-*