Submit a cohort
Generate many patients from one specification, cared for in one fictional health system. Returns at once with the cohort, queued; poll it with GET /v1/cohorts/{cohort_id}.
Authentication
An API key, or an OAuth 2.0 access token, with the synth:process scope.
Headers
Optional, recommended: retrying with the same key and body returns the original job (200) instead of creating, and charging for, a second one. Per organization, for 24 hours. 1-128 of [A-Za-z0-9_.:-].
Request
Conditions given to an exact share of the patients (rounded); others still arise naturally.
Only these document types. Chosen per patient if omitted. To get normal charts that also have certain types, use include_document_types instead.
Make sure every patient’s chart has at least one document of each of these types; the rest are the usual mix. At most documents_per_patient.min types.
Each patient’s fhir bundle, notes, and pdfs (as for a job), and bulk: the whole cohort as FHIR Bulk Data (NDJSON).
Random seed: the same request gives the same cohort.
Response
Idempotent retry: the cohort already created with this Idempotency-Key
Most documents the cohort can produce (size x documents per patient).

