Google Cloud · serverless containers

Cloud Run cost, second by second.

Turn requests, request duration, vCPU and memory into a monthly Cloud Run bill, with the free tier applied.

Request-based cost = (vCPU-seconds − 180,000) × $0.000024 + (GiB-seconds − 360,000) × $0.0000025 + (requests − 2M) × $0.40 per million. Time is rounded up to the nearest 100 ms. Rates are us-central1 (Tier 1).

Runs in your browserList prices retrieved October 2026

Advertisement

Your usage

Request-based inputs
Instance-based inputs
Rates (editable, us-central1 Tier 1)

Estimate

Method and limits

In request-based billing, billable time per request is the duration rounded up to the next 100 ms. Instance-seconds are requests × billed seconds ÷ concurrency, then multiplied by vCPU and GiB to get vCPU-seconds and GiB-seconds. Concurrency above 1 assumes instances are kept full, so treat it as a best case. Minimum instances add idle time, charged at Google’s idle rates ($0.0000025 per vCPU-second and per GiB-second), for the part of the month they are not serving requests. Instance-based billing charges every second an instance is allocated, with no per-request fee.

Not included: Cloud Run networking and egress (see the GCP egress calculator), build, Artifact Registry, logging, committed-use discounts, other regions, GPUs and Cloud Run functions.

Sources, retrieved October 2026: Cloud Run pricing (us-central1 (Tier 1): request-based $0.000024 per vCPU-second and $0.0000025 per GiB-second of active time, $0.40 per million requests; instance-based $0.000018 per vCPU-second and $0.000002 per GiB-second; free tier 180,000 vCPU-seconds, 360,000 GiB-seconds and 2 million requests (request-based) or 240,000 vCPU-seconds and 450,000 GiB-seconds (instance-based); usage rounded up to the nearest 100 ms). Estimates only: this is not a provider quote, and list prices change. Check the linked pages and your own account before committing spend.

Frequently asked questions

How is Cloud Run billed?

Google lists two modes. Request-based billing charges CPU and memory while a request is being handled, plus a per-million request fee. Instance-based billing charges CPU and memory for the full time an instance is allocated, with no request fee. Usage is rounded up to the nearest 100 ms.

What is the Cloud Run free tier?

For request-based billing Google lists 180,000 vCPU-seconds, 360,000 GiB-seconds and 2 million requests free each month, applied using us-central1 pricing. Instance-based billing lists 240,000 vCPU-seconds and 450,000 GiB-seconds. The tool applies them when the checkbox is on.

Does the calculator include egress?

No. Outbound internet traffic from Cloud Run is billed under Google Cloud network pricing, so use the GCP egress cost calculator and add the two.

Why does concurrency lower the bill?

When an instance handles several requests at once they share its CPU and memory, so instance-seconds fall. The tool divides billed time by concurrency, which is only accurate when instances stay busy.

Advertisement
Advertisement