Python services dynamically specify cron schedules

Introduced 2026-04-15 · score 0.91

HOW TO USE IT

When a Python cron service in `experimentalServices` needs multiple schedules, set `schedule` to `"<dynamic>"`. At build time Vercel imports and runs its entrypoint, which must return a list of `(entrypoint, cron_schedule)` pairs.

{
  "experimentalServices": {
    "scheduler": {
      "type": "cron",
      "entrypoint": "jobs.registry:crontab",
      "schedule": "<dynamic>"
    }
  }
}

WHY THIS HELPS

Configure multiple cron jobs for a Python service using schedules determined by its entrypoint.

REQUIREMENTS AND LIMITATIONS

Version: 51.4.0
experimentalServices.scheduler.type (required): Set to `"cron"` for a cron service.
experimentalServices.scheduler.entrypoint (required): Python entrypoint that Vercel imports and runs at build time; it must return a list of `(entrypoint, cron_schedule)` pairs.
experimentalServices.scheduler.schedule (required): Set to `"<dynamic>"` to use schedules supplied by the entrypoint.
Prerequisite: To deploy a project using `experimentalServices`, set its framework setting to Services and include `experimentalServices` in `vercel.json`. Services are in private beta and require access granted by request.
The behavior applies when the schedule for a cron service in `experimentalServices` is `"<dynamic>"`.
The entrypoint function must return a list of pairs of `(entrypoint, cron_schedule)`.
The entrypoint is imported and run at build time.
Docs conflict: Current docs describe the legacy `experimentalServices` model as available while migrating and say new projects should use the `services` model; they do not document dynamic Python cron schedules. https://vercel.com/docs/services/experimental

FIRST-PARTY EVIDENCE

Check the installed version and the source before using it. Entry JSON

Report a problem with this card

Tools change faster than your agent’s training data.

Changes

Python services dynamically specify cron schedules