📄 Parse Only Selected Pages of a PDF

Page processing lets you choose which pages of an incoming PDF Airparser should process. The rest are discarded before parsing — and before any document split.

This is useful when your PDFs contain pages you never need, such as blank backs of scanned sheets, cover pages, or appendices. Discarding them early reduces noise in the extracted data and saves parsing credits.

How to configure it

Go to Inbox Settings → General and scroll to the Page processing section.

Choose one of four modes:

• All pages — default. Every page is processed.

• Only odd pages: 1, 3, 5, … — keeps pages 1, 3, 5, and so on.

• Only even pages: 2, 4, 6, … — keeps pages 2, 4, 6, and so on.

• Page ranges — keeps only the pages you specify using the range syntax below.

Page ranges syntax

Enter a comma-separated list of pages or ranges:

• 1 — first page only

• 2-5 — pages 2 through 5

• (1) — last page

• (3)-(1) — last three pages

• 1, 3-5, (1) — pages 1, 3, 4, 5, and the last page

Page numbers are 1-based. Ranges that fall entirely outside the document are ignored.

Credits

Page processing runs before the document is sent for AI extraction. Only the pages that are kept count toward your credit usage.

For example, if you receive a 10-page PDF but keep only pages 1–3, you are charged for 3 pages — not 10.

Interaction with Document split

Page processing always runs first, before any document split. The split then operates on the already-filtered PDF.

For example, if you enable Only even pages and Split every 1 page, Airparser first discards odd pages and then creates one document per remaining page.