Limits

Requests per key

Each API key can make 600 requests per minute, over a sliding minute. Past that, requests get a 429:

{ "code": "rate_limited", "error": "Too many requests for this API key. Try again in 12 seconds." }

The Retry-After header holds the seconds to wait. Several keys don't share this limit, but they do share the workspace's plan limits below.

Project creation

Creating projects is limited per workspace and per hour, by plan. Past it, Create a project answers 429 rate_limited with Retry-After:

Plan limits

Builds, live environments, build machines and custom domains follow your plan, through the API as through the dashboard. A request past one gets a 402 quota_exceeded (Errors) naming the limit. The full list and every plan's values are on Limits.

Two you meet often from a script:

• Concurrent builds. A deployment past maxConcurrentBuilds waits as queued instead of failing, until its turn. Only a full queue is a 402 (quota: queued_builds).
• Live environments. A deployment that would make one more environment live than your plan includes is refused (quota: active_environments).