gcpServerless compute

Google Cloud Run

Runs a container in response to requests or events, scaling to zero when idle. Google Cloud's closest counterpart to Lambda, with a container rather than a function as the unit of deployment.

Limits

Pricing model

Per vCPU-second and GiB-second of allocated resources plus per-request charges, billed only while a container instance is handling requests under the default request-based model. Instance-based billing charges for the whole instance lifetime instead.

Cross-provider equivalents

  • AWS Lambda

    How it differs: Cloud Run's request timeout reaches 60 minutes (3,600 seconds) against Lambda's hard 900-second ceiling, and it deploys a container image rather than a zip of function code — so work that Lambda forces you to split into orchestrated steps often stays a single Cloud Run request.