Estimate Azure DevOps Pipelines costs for MS-hosted and self-hosted parallel jobs, build minutes, and Artifacts storage.
Last verified: May 2026
Microsoft manages the agent infrastructure. 1 free parallel job with 1,800 min/month.
Additional jobs: $40/month each (unlimited minutes per job)
2 GB free per organization. Additional storage at $2/GB/month.
Microsoft-hosted agents are fully managed VMs that spin up fresh for each job. You get 1 free parallel job with 1,800 minutes/month. Additional parallel jobs cost $40/month each and include unlimited minutes. Self-hosted agents run on your own infrastructure. The first parallel job is free with no minute limits. Additional jobs cost $15/month each. You manage the agent machine (patching, scaling, availability).
Add parallel jobs when your team frequently queues builds behind each other. Each parallel job allows one pipeline run at a time. If you have 3 parallel jobs, 3 pipelines can execute simultaneously. For MS-hosted, the free tier's 1,800 minutes/month may also drive the need for additional jobs, since each additional job gets unlimited minutes.
Azure Artifacts provides 2 GB of free storage per organization for packages (NuGet, npm, Maven, Python, Universal). Beyond the free tier, storage costs $2/GB/month. To manage costs, configure retention policies to automatically clean up old packages and pipeline artifacts. Use upstream sources to cache external packages without counting against your storage.
GitHub Actions: 2,000 free minutes/month (public repos unlimited), $0.008/minute for Linux beyond free tier. GitLab CI/CD: 400 free compute minutes/month on shared runners. Jenkins: Free and open source, but you pay for infrastructure. CircleCI: 6,000 build credits/month free, pay-per-use beyond that. Azure DevOps is competitive for teams already in the Microsoft ecosystem.
Output will appear here...Your team's CI/CD bill on Azure DevOps Pipelines is $850/month for 15 Microsoft-hosted parallel jobs. The estimator models the alternative: 15 self-hosted parallel jobs ($225/month) + 5 Standard_D4s_v5 VMs ($580/month) = $805/month. Same parallelism, $45/month savings. But more importantly, self-hosted agents have unlimited build minutes — your build duration variability no longer affects the bill. You migrate and stop worrying about long Terraform plans burning minutes.
The Azure DevOps Pipeline Cost Estimator helps you estimate monthly costs for Azure DevOps Pipelines based on your CI/CD usage patterns. You can configure the number of parallel jobs, build minutes, agent types (Microsoft-hosted or self-hosted), and pipeline frequency to get a cost projection. The tool breaks down costs by hosted agent tier and shows how different configurations affect your monthly bill, helping teams budget for their DevOps workflows.
The estimator calculates monthly cost as: (parallel jobs × per-job-monthly fee) + (build minutes above free tier × per-minute rate). It supports both Microsoft-hosted and self-hosted modes, with the latter requiring you to also factor underlying VM costs (which the estimator surfaces as a separate line item from a representative VM SKU).
Microsoft-hosted parallel jobs cost $40/month each beyond the free tier. Self-hosted agents cost $15/month per parallel job — and you can run multiple builds on a single self-hosted agent. For organizations with >5 parallel jobs, self-hosted typically pays for itself within 2-3 months even after factoring VM costs.
Microsoft-hosted agent build minutes count even when builds queue waiting for an available agent. If your peak builds need 20 parallel jobs but you only have 10 paid, the queued builds still consume minutes once they start. Always size parallel jobs to your peak concurrent build count, not your average.
The 1,800 free minutes for private projects sound generous, but a typical 10-minute build × 30 builds/day × 22 working days = 6,600 minutes/month — well over the free tier. Always estimate against actual build cadence, not just the included minutes.
Was this tool helpful?
Disclaimer: This tool runs entirely in your browser. No data is sent to our servers. Always verify outputs before using them in production. AWS, Azure, and GCP are trademarks of their respective owners.