Convert PDF Page to JPEG

This Vision endpoint allows you to convert individual PDF pages into .jpeg format, enabling you to send images to GPT-4.o, which cannot perform OCR directly from PDF files.

Important! If you need to convert multiple PDF pages, you must make a separate API request for each page you want to convert.

Note: This feature is available in DRUID 8.2 and higher. To use this Druid Vision service, you must have at least one Identity Vision API service enabled for your tenant. If not, please contact Druid Tech Support.

Request

Method	POST
API Resource URL (Endpoint)	*.druidplatform.com/api/document-processing/convert-pdf-page-to-jpeg
Params	Enter the pageNumber parameter and specify the PDF page number you want to convert to a .jpeg.
Authorization	Select Bearer Token. If you already have a Druid Vision Token generated for an Identity API service in the DRUID Portal, provide that token. For details on obtaining the token, refer to Druid Vision. If you don't have a token, generate one using the Authentication endpoint from Webhooks. For more information, see Webhooks.
Body	“file”: <attached document file> (file attachment) Supported formats: pdf

Response

The response returns the specified PDF page as a .jpeg file.

To capture the .jpeg file, use the Download File integration. You can then pass the file to GPT-4.o within your knowledge base-specific flows.