Document Translation API

When to use this API

Document Translation API is built for production workflows where text translation alone is not enough. APIRAPTOR extracts translatable units from the original file, translates only controlled segments, and reconstructs the final document in its original structure as closely as technically possible for the given format.

• You need multilingual contract workflows without rebuilding document layout manually.
• You process offers, manuals, sales documents and internal files that must preserve structure.
• You need one API flow that handles text PDFs, scanned PDFs, office documents and plain-text formats.

Typical use cases

• Translation of agreements and legal documents.
• Translation of commercial offers and sales materials.
• Translation of operating manuals and technical instructions.
• Translation of scanned paperwork and PDF archives.
• Translation of spreadsheets and HTML/TXT/Markdown content without exposing internal segment payloads.

Billing and APIRAPTOR token model

Document Translation usage is metered per translated page through the `translated_page` service billing event.

• Tokens are the APIRAPTOR communication unit: 1 token = 100 subtokens.
• Trial availability, token amount and duration come from Billing configuration.
• All services are available from day one; Document Translation is not premium-gated per plan.
• This is service billing event metering, not ordinary API-call billing.

Translation quality boundary

Document Translation preserves structure and protected technical tokens as closely as the source format allows, but translation and layout reconstruction remain reviewable outputs.

• Layout preservation depends on input format, scan quality, embedded objects, tables and OCR fallback quality.
• Translated content should be reviewed before legal, regulated or customer-critical publication.
• The service does not provide certified, sworn or formal legal translation and does not guarantee perfect translation.
• The service translates controlled segments and does not return raw provider output or internal segment diagnostics in public responses.

Sync and async usage

Sync and async requests use the same canonical Document Translation POST endpoint.

• Sync uses POST /v1/document-translation with mode=sync or the configured default mode.
• Async create uses POST /v1/document-translation with mode=async.
• Jobs endpoints are used for status/result metadata, not job creation. File artifacts are returned as file_id values for authenticated API download or signed URLs when direct download is enabled, and generated file artifacts are deleted automatically no later than 48 hours after creation.