Pricing

This page provides pricing information for Cloud Run (fully managed) and Cloud Run for Anthos on Google Cloud.

With regard to deploying container images from Container Registry, as documented in Cloud Storage network pricing, you are not charged if you deploy to a region located on the same continent as your container image. For example, deployments to a Cloud Run (fully managed) service in us-central1 or us-east1 from us.gcr.io or gcr.io are free of charge.

Cloud Run (fully managed) pricing

Cloud Run (fully managed) charges you only for the resources you use, rounded up to the nearest 100 millisecond. Note that each of these resources have free quota. Your total Cloud Run (fully managed) bill will be the sum of the resources in the pricing table.

When setting concurrency higher than one request at a time, multiple requests can share the allocated CPU and memory.

Pricing calculator

You can use the Google Cloud Platform Pricing Calculator to estimate the cost of using Cloud Run (fully managed).

Pricing tables

The following pricing tables use the GiB-second unit. A GiB-second means for example running a 1 gibibyte instance for 1 second, or running a 256 mebibyte instance for 4 seconds. The same principle applies for the vCPU-second unit.

Tier CPU Memory Requests Networking
Free First 180,000 vCPU-seconds free per month First 360,000 GiB-seconds free per month 2 million requests free per month 1 GiB free egress within North America per month
1 $0.00002400 / vCPU-second beyond free quota
($0.00000250 if idle*)
$0.00000250 / GiB-second beyond free quota
($0.00000250 if idle*)
$0.40 / million requests beyond free quota Google Cloud Network Premium tier pricing beyond free quota.
2 $0.00003360 / vCPU-second beyond free quota
($0.00000350 if idle*)
$0.00000350 / GiB-second beyond free quota
($0.00000350 if idle*)
$0.40 / million requests beyond free quota Google Cloud Network Premium tier pricing beyond free quota.

If you pay in a currency other than USD, the prices listed in your currency on Cloud Platform SKUs apply.

* idle refers to instances kept warm using minimum instances.

Subject to Tier 1 pricing

  • asia-east1 (Taiwan)
  • asia-northeast1 (Tokyo)
  • asia-northeast2 (Osaka)
  • europe-north1 (Finland)
  • europe-west1 (Belgium)
  • europe-west4 (Netherlands)
  • us-central1 (Iowa)
  • us-east1 (South Carolina)
  • us-east4 (Northern Virginia)
  • us-west1 (Oregon)

Subject to Tier 2 pricing

  • asia-east2 (Hong Kong)
  • asia-northeast3 (Seoul, South Korea)
  • asia-southeast1 (Singapore)
  • asia-southeast2 (Jakarta)
  • asia-south1 (Mumbai, India)
  • australia-southeast1 (Sydney)
  • europe-west2 (London, UK)
  • europe-west3 (Frankfurt, Germany)
  • europe-west6 (Zurich, Switzerland)
  • northamerica-northeast1 (Montreal)
  • southamerica-east1 (Sao Paulo, Brazil)

Note that it is not possible to use the domain mapping feature of Cloud Run (fully managed) for services in these regions:

  • asia-east2
  • asia-northeast2
  • asia-northeast3
  • asia-southeast1
  • asia-southeast2
  • asia-south1
  • australia-southeast1
  • europe-west2
  • europe-west3
  • europe-west6
  • northamerica-northeast1
  • southamerica-east1
You can use Cloud Load Balancing with a serverless NEG to map a custom domain to Cloud Run (fully managed) services in these regions.

The free quota is aggregated across projects by billing account and resets every month; you are billed only for usage past the free quota. The free tier is applied as a spending based discount using Tier 1 pricing.

Billable Time

For a given container instance, billable time occurs when:

  • The container instance is starting
  • At least one request is being processed by the container instance

You are billed only for the CPU and memory allocated while a request is active on a container instance, rounded up to the nearest 100 milliseconds, unless you set a minimum number of container instances.

If a container instance receives many requests at the same time, billable time begins with the start of the first request and ends at the end of the last request, as shown in the following diagram:

Billable time

The billable time aggregated from all Cloud Run (fully managed) container instances is exposed as a Cloud Monitoring metric. See container/billable_instance_time metric for more details.

Eventarc pricing (preview)

Eventarc has the following billable components:

  • Events Core
  • The Pub/Sub transport layer

    Eventarc architecture

Events from Google Cloud sources using Cloud Audit Logs use Events Core and Pub/Sub as the transport layer. Events from custom sources only use Pub/Sub as the transport layer.

The total Eventarc billable charge is the combined charge of Events Core and Pub/Sub.

Events using Cloud Audit Logs incur Cloud Logging charges and event receiver services on Cloud Run (fully managed) incur Cloud Run (fully managed) charges.

Events Core pricing

Events from Google Cloud sources are charged at $0/million events published.

Transport layer pricing

Usage of the Pub/Sub transport layer is billed at standard Pub/Sub rates.

Billing examples

These examples assume events using Cloud Audit Logs are typically 2KiB.

Google Cloud event source example

Scenario: A Cloud Run (fully managed) service receiving one 2 KiB per second for a month (2,592,000 events) at a rate of 2 KBps from a Google Cloud source such as Cloud Storage using Cloud Audit Logs.

Usage of Events Core: US$0/million events published for 2,592,000 events. The change is US$0.

Usage of the Pub/Sub transport layer: Assuming one Pub/Sub topic with one subscriber that is consuming all generated events.

Pub/Sub has a total transmission rate of 4 KBps (2 KBps ingress and 2 KBps egress). Over 30 days this is 0.0095 TiB of data. The first 10GB is part of the free tier, the remaining billable data volume is 0 TiB, and the charge is US$0.

Total charge: US$0

Custom event source example

Scenario: A Cloud Run (fully managed) service receiving one 1024KiB event every second for a month (2,592,000 events) at a rate of 1 MBps from a custom source.

Usage of Events Core: not applicable.

Usage of the Pub/Sub transport layer: Assuming one Pub/Sub topic with one subscription and the subscriber is consuming all generated events.

Pub/Sub has a transmission rate of 2 MBps (1 MBps ingress and 1 MBps egress). Over 30 days this is 4.894 TiB of data. The first 10GB is part of the free tier, the remaining billable data volume is 4.884 TiB, when charged at US$40 per TiB, generates a billable charge of US$195.36.

Total charge: US$195.36

Cloud Run for Anthos on Google Cloud pricing

Cloud Run for Anthos on Google Cloud is an add-on for Google Kubernetes Engine. Workloads running in a cluster are included in Google Kubernetes Engine pricing.

The Generally Available (GA) version of Cloud Run for Anthos on Google Cloud is a feature of the Anthos platform. Use of Cloud Run for Anthos on Google Cloud is included in the Anthos subscription. Contact your sales representative to enroll in Anthos.

Events for Cloud Run for Anthos pricing (beta)

Events for Cloud Run for Anthos is an add-on for Cloud Run for Anthos on Google Cloud. Workloads running in a cluster are included in Google Kubernetes Engine pricing. Events for Cloud Run for Anthos makes use of other Google Cloud services that generate billable cost.

Service Use
Google Kubernetes Engine Container pods for the eventing components. The compute resources used in these pods will generate billable cost.
Pub/Sub Pub/Sub is used to transport events. The generation of Pub/Sub topics will generate billable cost.
Cloud Logging Cloud Logging is used to generate audit and service logs. The generation of these logs will generate billable cost.
Cloud Trace Cloud Trace is used to track event delivery from source to receiver. Cloud Trace ingestion spans will generate billable cost.

Cloud Run for Anthos on Google Cloud free trial period

Google is offering a trial period during which the product remains free for new and existing customers. The free trial period ends after March 31, 2021. Underlying charges for using the GKE, Pub/Sub, Cloud Trace, and Cloud Logging infrastructure still apply.