開発者向けドキュメントGet started
Rate limits & limits
Request rates, job size, concurrency and retention.
Defaults below. GET /v1/limits returns the values for your organization.
| Limit | Value |
|---|---|
| Requests per minute / hour / day | 60 / 1,000 / 5,000 |
Item units per job (sum of quantity) | 2,000 |
| Vehicles per job (incl. added ones) | 20 |
| Compute budget per job | 90 s |
| Jobs computing at once | 1 |
| Jobs queued + running | 20 |
| Vehicle types | truck, trailer, container, pallet |
| Request body | 10 MB |
?wait | 60 s |
| Retention of bodies, results, PDF links | 14 days |
Rate limits
Counted per minute, hour and day, for each key and for the organization. The strictest wins.
| Header | Meaning |
|---|---|
X-RateLimit-Limit | Limit of the tightest window |
X-RateLimit-Remaining | Requests left in it |
X-RateLimit-Reset | Unix time it resets |
Retry-After | On 429: seconds to wait |
Over the limit: 429 rate_limited. More than 60 failed authentications per minute from one IP also answer 429.
Jobs
Jobs over the concurrency limit wait as queued. Past 20 queued + running jobs, POST /v1/pack answers 429 too_many_jobs.
Retention
Request and result bodies, and PDF links, are deleted after 14 days (expires_at on every job). Status, counts and cost are kept.