AnyConvert
← All docs

AI Bundle: Tiles with Text

Send a PDF or DWG drawing, get back a zip of high-resolution tile images, each paired with a hints file listing the exact words positioned on that tile. Built for handing a dense drawing to an AI model one region at a time.

See the AI Bundle: Tiles with Text tool page, or the docs index for authentication, the shared request/webhook contract, and the generic error table.

Request

POST /webhooks/trigger — same endpoint and fields as every tool, with tool="pdf-tiles-with-text" set explicitly (it's not the default).

curl
curl -X POST https://api.anyconvert.app/webhooks/trigger \
  -H "x-api-key: your_api_key_here" \
  -F "file=@pid-sheet.pdf" \
  -F "fileId=drawing-001" \
  -F "referenceId=reference-001" \
  -F "tool=pdf-tiles-with-text"

Options

An optional options multipart field carrying a JSON object, validated synchronously before a credit is charged — a bad value is rejected with 400, not charged and not queued.

profilestring, optionalDefault "medium". Tile accuracy/zoom profile: "low", "medium", "high", or "max".
pagesstring, optionalDefault "all"; the page-selection grammar.
textEnginestring, optionalDefault "auto". For PDF input: "auto", "text-layer", or "ocr". DWG input always uses exact entity text.
minConfidencenumber, optionalDefault 0. Words below it stay in the list, flagged lowConfidence — never removed.
requireTextboolean, optionalDefault false. If true and no text is found anywhere, the job fails instead of delivering tiles with a warning.
includeLowResOverviewboolean, optionalDefault true — add a single low-resolution full-page overview image with all word boxes, for a first pass before the tiles.

Response

The webhook delivers a .zip: manifest.json (every tile's location, size, and status), text.json (the full underlying text-extraction result), one PNG per tile under tiles/, one hints JSON per tile under hints/ listing that tile's words with their pixel position, source, and confidence, and a README.txt explaining how to use the bundle.

Pricing

Billed per page, not per document — 1 credit per page, covering the tile images and the text extraction for that page together (and the DWG-to-PDF conversion step, when the input is a DWG).

Failure reasons

Delivered via the webhook as outcome=failure with a reason below — see the docs index for the full delivery shape.

invalid_optionsA bad value for profile, textEngine, pages, or another option.
encrypted_fileThe input PDF is password protected.
dwg_conversion_failedThe DWG input could not be converted ahead of tiling.
no_extractable_textNo text was found anywhere in the document and requireText was set.
invalid_fileThe file could not be read, or was not the format this tool expects.
render_failedSomething failed during conversion that wasn’t a problem with the input itself.
timeoutThe job did not finish within its time budget.
internal_errorAn unexpected error on our side.
AnyConvert
A simple conversion API for teams that need documents turned into the right format, automatically.
Tools
PDF to WordMerge PDFJPG to PDFSign PDFEdit PDFCompress PDFExtract Text with CoordinatesView all tools →
Product
DocsPricing
Legal
Terms of ServicePrivacy PolicyRefund Policy
Support
support@anyconvert.app
© 2026 AnyConvert. All rights reserved.