Vercel Functions Cost Calculator
Pick a Fluid region (default Washington D.C. iad1), enter invocations, active CPU ms per request, memory GB and instance-alive ms (incl. I/O). See CPU $, memory $, invocation $ and total on Pro on-demand with no Hobby/Pro usage credit applied. Enterprise and non-Fluid legacy skipped. Estimates only; vercel.com/docs/functions/usage-and-pricing prevails.
Math runs in your browser. Nothing is uploaded.
Monthly total
—
CPU · memory · invocations
—
Breakdown
—
Vercel Fluid bills Active CPU (code execution only, not I/O wait) + Provisioned Memory (GB-hours while the instance is alive, incl. I/O) + Invocations. Default region iad1: Active CPU $0.128/hour, Provisioned Memory $0.0106/GB-hour. Invocations Pro on-demand: $0.60 per million. MVP default = Pro on-demand with no Hobby credit / Pro monthly usage credit applied (footnote: Pro credit may offset; Hobby includes 4 CPU-hrs / 360 GB-hrs / 1M invocations). Smoke A (iad1): 15M inv, 10 ms active, 1 GB, 50 ms alive → CPU ≈ $5.333 + mem ≈ $2.208 + inv $9.00 ≈ $16.54. Enterprise and non-Fluid legacy skipped.
Estimates only. Official list prices as of 2026-11-03 (Asia/Shanghai); the provider's pricing page prevails. Source: Vercel Functions usage and pricing
Serverless Compute Cost Comparison · Cloudflare Workers Cost Calculator · AWS Lambda Cost Calculator · CDN Bandwidth Cost Comparison
More in this group
- Serverless Compute Cost Comparison Workers vs Lambda vs Vercel Fluid monthly compute $.
- Cloudflare Workers Cost Calculator Workers Paid requests + CPU-ms → monthly $.
- AWS Lambda Cost Calculator Lambda Event us-east-1 x86 requests + GB-s → monthly $.
- CDN Bandwidth Cost Comparison Bunny vs Cloudflare vs CloudFront vs Fastly monthly CDN $.
- Object Storage Cost Comparison R2 vs S3 vs B2 storage + egress monthly bill.
- PaaS Hosting Cost Comparison Railway · Render · Fly · Netlify · Heroku monthly $.
How it works
- How does Vercel Fluid pricing work?
- You pay Active CPU (code execution only), Provisioned Memory (GB-hours while the instance is alive, including I/O waits) and Invocations ($0.60 per million on Pro on-demand). Rates vary by region.
- What is the iad1 Scenario A total?
- 15M invocations, 10 ms active CPU, 1 GB memory, 50 ms alive: CPU hours = 41.6667 × $0.128 ≈ $5.333; mem GB-hrs = 208.333 × $0.0106 ≈ $2.208; invocations = 15 × $0.60 = $9.00; total ≈ $16.54.
- Are Hobby or Pro credits applied?
- MVP default is Pro on-demand with no Hobby credit and no Pro monthly usage credit applied. Footnote: Hobby includes 4 Active CPU hours, 360 GB-hrs memory and 1M invocations; Pro credit may offset real bills.
- Why does memory keep billing during I/O?
- Provisioned Memory is billed for the entire instance lifetime while requests are in flight. Active CPU pauses during I/O; memory does not — which matters on DB/API-heavy workloads.
- Which regions are included?
- Published Fluid regions including iad1/cle1/pdx1 ($0.128/$0.0106), fra1 ($0.184/$0.0152), sin1 ($0.160/$0.0133) and the other rows from Vercel's regional table (as of 2026-11-03).