Skip to main content
GET
Rate limit tier: general, 600 requests/min. See Rate limits. No credits are billed. A workflow with a splitter node carves one uploaded packet into sub-documents — the splits[] a run’s results carry. This endpoint returns one of them as a PDF. The document is built from the packet’s source files when you ask for it and is never stored, so there is no artifact to expire and no second URL to follow: the bytes come back on this request.

Finding the id

Read it off a processed run’s results (Get a run). An id is present when, and only when, it addresses exactly one sub-document: A partitioned category is several sub-documents and no single one of them, so it has no id of its own. Download each partition instead.

Caching

The response carries an ETag and Cache-Control: private. The tag is opaque — store the one you were given and send it back verbatim in If-None-Match, rather than building it from the split id. A split’s pages never change (a new run of the same workflow mints new split ids), so a still-current split answers 304 Not Modified without transferring the document again.

Errors

Authorizations

Authorization
string
header
required

API key issued from app.anyformat.ai/api-key. Send as Authorization: Bearer <key>.

Headers

if-none-match
string | null

Path Parameters

split_id
string
required

Response

The split's pages as a PDF.