Requests vs compute
Short, light functions are request-bound, long or memory-heavy ones flip to compute-bound.
Home / Dev & AI / Cloud Cost / Google Cloud Run Cost Calculator
Cloud Cost
Google Cloud Run Cost Calculator
Estimate your Cloud Run bill. Enter requests, vCPU, memory and duration, and see request and compute costs with the free tier applied and a growth projection.
Estimated monthly cost
$0
Request cost—
Compute cost (GB-seconds)—
GB-seconds used—
Cost per 1M invocations—
Annual estimate—
Cost as invocations grow
Projected monthly spend at 2x, 5x and 10x volume.
What the Google Cloud Run Cost Calculator does
Cloud Run runs containers serverlessly, scaling to zero when idle so you pay only while requests are being served. Billing has three parts: a per request fee, vCPU time and memory time, both measured per second while your container handles traffic. A free tier covers the first two million requests and a large slice of compute each month.
This Cloud Run cost calculator estimates compute from your allocated vCPU, memory and average request duration, applies the request and compute rates, and subtracts the free tier. Because you are billed only during request processing on the default model, short, efficient handlers keep costs very low.
Heads up on pricing. The rates built into this tool are public list estimates for July 2026 and can change without notice. Providers also offer batch discounts, prompt caching and volume tiers that lower real costs, so treat the output as a planning estimate, not a quote.
The formula
compute (GB-seconds) = requests × (duration ÷ 1000) × (memory ÷ 1024)
cost = (billable requests ÷ 1M × $0.40) + (compute × price per GB-second) + vCPU time
cost = (billable requests ÷ 1M × $0.40) + (compute × price per GB-second) + vCPU time
Worked example
For 10,000,000 requests at 200 ms and 512 MB, memory time is 10,000,000 × 0.2 × 0.5 = 1,000,000 GB-seconds. After the free tier, requests cost 8 × $0.40 = $3.20 and compute a few dollars more, keeping a light service around $5 to $10 a month.
Ways to lower your cloud costs
- Right size resources. Pick the smallest instance, memory or capacity that still meets your performance target.
- Commit to steady load. Reserved instances, savings plans and committed use discounts cut rates sharply for predictable workloads.
- Use serverless for spiky traffic. Scale-to-zero services mean you pay nothing while there are no requests.
- Tier cold storage. Move rarely accessed data to cheaper classes such as Glacier, Coldline or Archive.
- Cut data transfer. A CDN and keeping traffic within a single region reduce the egress charges that quietly grow with scale.
Frequently asked questions
How is Cloud Run priced?
By requests at about $0.40 per million, plus vCPU and memory time per second while serving. A generous free tier covers the first 2 million requests and much of the compute each month.
Does Cloud Run scale to zero?
Yes. With the default request based billing, idle services cost nothing because no vCPU or memory time is billed when there are no requests.
Is Cloud Run cheaper than a VM?
For spiky or low traffic, yes, since it scales to zero. For steady high load, a committed Compute Engine VM can be cheaper per request.
How do I lower Cloud Run costs?
Reduce allocated memory and vCPU to the minimum that performs well, keep handlers fast, and set concurrency higher so each instance serves more requests.
Are these rates current?
They are July 2026 estimates. Confirm current vCPU, memory and request pricing on the Google Cloud pricing page for your region.