specification: API Commons Rate Limits
specificationVersion: '0.1'
schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/RateLimits
provider: Azure Pipelines
providerId: microsoft-azure-pipelines
created: '2026-05-04'
# Provenance stamped 2026-08-11: this artifact was written by the API Evangelist
# bulk sweep dated 2026-05-04, not harvested from the provider. See roadmap#35.
method: generated
modified: '2026-05-05'
reconciled: true
tags:
  - Automation
  - Build
  - CI/CD
  - Deployment
  - DevOps
  - Pipelines
  - Rate Limiting
description: Azure DevOps Services (which hosts Azure Pipelines) applies a per-user / per-identity
  Throughput Service Threshold (TSTU / RCU) sliding-window rate limit. Each REST API call
  consumes a number of TSTUs based on cost; each user has 200 TSTUs per 5-minute sliding window.
  Throttled requests return HTTP 429 with Retry-After. Concurrency is capped not by API rate
  but by purchased parallel jobs (default 25 for Microsoft-hosted on new orgs, raise via support).
  Per-user TSTU limits cannot be raised — clients must throttle.
sources:
  - https://learn.microsoft.com/en-us/azure/devops/integrate/concepts/rate-limits
  - https://learn.microsoft.com/en-us/azure/devops/pipelines/licensing/concurrent-jobs
headers:
  retryAfter: Retry-After
  delay: X-RateLimit-Delay
  limit: X-RateLimit-Limit
  remaining: X-RateLimit-Remaining
  reset: X-RateLimit-Reset
  resource: X-RateLimit-Resource
responseCodes:
  throttled: 429
  delayedNotBlocked: 200
limits:
  - name: Per-user TSTU sliding window
    scope: user
    metric: tstu_per_5_minutes
    limit: 200
    timeFrame: minute
    notes: Throughput measured in Throughput Units (TSTUs). 200 per 5-minute sliding window
      per user. Heavy queries cost more TSTUs; light reads can cost <1.
  - name: Microsoft-hosted parallel job concurrency
    scope: organization
    metric: concurrent_jobs
    limit: 25
    notes: New organizations are capped at 25 Microsoft-hosted parallel jobs until a support
      ticket raises the limit; subject to regional capacity.
  - name: Microsoft-hosted job duration (paid)
    scope: pipeline-run
    metric: minutes_per_run
    limit: 360
    timeFrame: minute
    notes: Each paid Microsoft-hosted job has a 6-hour per-run timeout.
  - name: Microsoft-hosted job duration (free private)
    scope: pipeline-run
    metric: minutes_per_run
    limit: 60
    timeFrame: minute
    notes: Free private projects cap individual runs at 60 minutes.
  - name: Self-hosted job duration
    scope: pipeline-run
    metric: minutes_per_run
    limit: -1
    notes: No per-run timeout on self-hosted parallel jobs.
policies:
  - name: TSTU cost weighting
    description: API calls cost differing TSTU values based on database/CPU expense. Inexpensive
      reads may cost <1 TSTU; expensive list/query operations may cost 5+.
  - name: Delay before block
    description: Before returning 429, the service may inject a delay (X-RateLimit-Delay) header
      to slow the client without blocking. Persistent abuse upgrades to 429.
  - name: Spread automation across identities
    description: Build agents under PAT or service principal share the same per-identity quota;
      use multiple service principals to fan out automation.
  - name: Exponential backoff
    description: Honor Retry-After. Implement exponential backoff with jitter for 429 responses.
  - name: No support increase for TSTU
    description: Per-user TSTU limits cannot be raised by support. Design clients to respect
      headers and stay under limit.
maintainers:
  - FN: Kin Lane
    email: kin@apievangelist.com
    url: https://azure.microsoft.com/services/devops/pipelines/