Extract up to 10 video URLs in one call

Fans out to the same engine as /v1/extract with bounded concurrency. Each url is SSRF-validated and processed independently: a failing url returns { url, ok:false, error } without failing the batch. Billed per SUCCESSFUL item at that item's own cost; failed items are not billed. Auth, balance and rate-limit are enforced once at the batch level (one rate token per call). Frames/text_overlay need a per-url frames spec, which batch does not accept — request those via /v1/extract.

post/v1/batch

Request

  • Base URL: https://framefetch.net
  • URL: https://framefetch.net/v1/batch
  • Auth: HTTP bearer

Request body

urlsstring[] required

1–10 supported-platform video URLs.

fieldsstring[]

Applied to every url. digest, delta, structured, audio_digest, comments and comment_sentiment work in batch (audio_digest always uses the default "alloy" voice — batch does not accept a top-level voice or translate param; comments always uses the default cap of 30 — batch does not accept a top-level comments_cap); frames/text_overlay need a per-url frames spec (not accepted here) — use /v1/extract for those.

Response

Per-url results, in input order

Changes

No recorded changes to this operation across all 1 revision of this API.