AWS Lambda vs Fargate Cost Calculator

Compare Lambda (requests + GB-seconds) vs Fargate (vCPU-hours + GB-hours) using your own pricing inputs. See baseline vs peak differences in a compute-only comparison.

Maintained by CloudCostKit Editorial Team. Last updated: 2026-01-28. Editorial policy and methodology.

Best next steps

Use this calculator for the first estimate, then validate the answer with the closest guide or companion tool.

Guide: Lambda vs Fargate cost Guide: Lambda pricing Guide: Fargate pricing Lambda cost calculator

Lambda inputs

Invocations (per month)
Avg 19.04 req/sec.
Avg duration (ms)
Memory (MB)
0.06 GB-sec per invocation.
Price ($ / 1M requests)
Price ($ / GB-second)
Deducts 1M requests + 400k GB-sec.

Fargate inputs

Running tasks (average)
vCPU per task
Memory (GB) per task
Hours/day
Days/month
Use 30.4 for an average month.
Monthly hours: 730
Price ($ / vCPU-hour)
Price ($ / GB-hour)

Comparison

Lambda monthly total
$53.13
Fargate monthly total
$54.03
Cheaper (compute-only)
Lambda
Difference (Fargate - Lambda)
$0.90 (169% vs Lambda)
Lambda cost per 1M
$1.08
Fargate cost per task
$18.01 (730 hr)
This compares compute-only pricing. Add logs, load balancers, NAT/egress, and retries for a full model.
Fargate uses 30.4 days/month and 24 hours/day.

Breakdown (sanity checks)

Lambda billable invocations
49,000,000
Lambda billable GB-seconds
2,600,000
Fargate vCPU-hours
1,094
Fargate GB-hours
2,189

Scenario presets

Reset

Lambda vs Fargate is a burst-shape question first

This comparison works best when you frame the workload by shape before you frame it by price. Lambda is strongest when execution is bursty and idle time is real. Fargate becomes stronger as work turns into steady, always-on capacity that keeps tasks busy for much of the month.

  • Lambda lens: request count, duration, and memory per execution.
  • Fargate lens: average running tasks, task size, and uptime window.
  • Boundary line: background work and persistent concurrency often move the decision toward Fargate.

What usually distorts this comparison

  • Cold-start or peak duration is used as the average duration for every Lambda invocation.
  • Provisioned concurrency or steady background traffic is ignored, making Lambda look cheaper than real operation.
  • Fargate is modeled as always-on even when the service only runs for specific windows.
  • Retries, logs, and networking are forgotten even though they can move the real platform decision.

When a hybrid model is more honest than either extreme

  • Use Lambda for burst and event fan-out when requests are spiky and idle gaps are meaningful.
  • Use Fargate for steady baselines, warm state, or long-running workers that would keep Lambda active continuously.
  • Model the baseline and the burst separately if the system genuinely has both shapes.
  • Do not force a binary answer when the workload already behaves like two different systems.

Baseline vs spike-driven runtime scenarios

Scenario Lambda requests Fargate tasks Notes
Baseline Expected Average Normal traffic
Peak High High Launch or incident

How to review the first real month

  • Check Lambda requests and GB-seconds against billing and verify that the modeled average duration matches observed execution data.
  • Check Fargate average task count and runtime against ECS history instead of against deployment peaks.

Next steps

Guides All calculators Guide: Lambda vs Fargate cost Guide: Lambda pricing Guide: Fargate pricing Lambda cost calculator

Example scenario

  • 50M requests/month at 120 ms and 512 MB vs 3 tasks running 24/7 at 0.5 vCPU and 1 GB.
  • For steady high-throughput workloads, Fargate often wins; for bursty spiky workloads, Lambda often wins.
  • Peak 180% scenario helps validate incident months.

Included

  • Lambda request cost and compute cost (GB-seconds) with optional free tier.
  • Fargate vCPU-hour and memory GB-hour compute costs from average running tasks.
  • Days/month input to align with billing cycles for Fargate.
  • Side-by-side monthly totals and differences.
  • Baseline vs peak scenario table for workload spikes.

Not included

  • Logs/metrics costs (often meaningful for services with high volume).
  • Networking costs (NAT, egress, cross-AZ/cross-region transfer).
  • Load balancers, retries/timeouts, and ancillary services.

How we calculate

  • Lambda: request cost + (duration x memory) GB-seconds x $/GB-second (minus free tier if enabled).
  • Fargate: tasks x vCPU x hours x $/vCPU-hour + tasks x memory(GB) x hours x $/GB-hour.
  • Hours = days per month x hours per day.
  • Compare the compute-only monthly totals, then add other line items in your overall model.

FAQ

Is this an official AWS quote?
No. This is a simplified estimate for planning. Verify region pricing, free tier, rounding rules, and service add-ons in AWS documentation.
Why does the winner change by workload shape?
Lambda pricing is execution-based and can be cheaper when traffic is bursty or idle often. Fargate pricing is capacity x time and can be cheaper for steady utilization and long-running compute.
What should I model next after compute?
Add logs, networking (egress/NAT/cross-AZ), and load balancers. For many services, these line items can be comparable to compute.

Related tools

Guide: Lambda vs Fargate cost Guide: Lambda pricing Guide: Fargate pricing Lambda cost calculator Fargate cost calculator Log cost Data egress cost

Related guides

S3 pricing: a practical model for storage, requests, egress, and replication
A practical S3 pricing guide: what to include (GB-month, requests, egress, replication) and how to estimate the key inputs without copying price tables.
CDN Cost Comparison Guide: Compare Pricing, Per-GB Rates, and Provider Trade-Offs
Compare CDN pricing across providers with a practical framework for bandwidth, requests, per-GB rates, regional mix, and origin egress. Built for CDN cost comparison and provider-decision workflows.
Cloud cost estimation checklist: build a model Google (and finance) will trust
A practical checklist to estimate cloud cost without missing major line items: requests, compute, storage, logs/metrics, and network transfer. Includes a worksheet template, validation steps, and the most common double-counting traps.
Copy storage pricing: what you pay for when data moves
A practical guide to pricing storage copy operations (cross-region copy, replication, backups) across S3-like object storage: transfer, requests, and extra storage.
Google Kubernetes Engine (GKE) pricing: nodes, networking, storage, and observability
GKE cost is not just nodes: include node pools, autoscaling, requests/limits (bin packing), load balancing/egress, storage, and logs/metrics. Includes a worked estimate template, pitfalls, and validation steps to keep clusters right-sized.
Lambda vs Fargate cost: a practical comparison (unit economics)
Compare Lambda vs Fargate cost with unit economics: cost per 1M requests (Lambda) versus average running tasks (Fargate), plus the non-compute line items that often dominate (logs, load balancers, transfer).

Disclaimer

Educational use only. Not legal, financial, or professional advice. Results are estimates based on the inputs and assumptions shown on this page. Verify pricing and limits with your providers and documentation.

Last updated: 2026-01-28. Reviewed against CloudCostKit methodology and current provider documentation. See the Editorial Policy .