Pictures and documents
Pictures and files become words people can search for. These calls tell what a picture shows, read the text printed in it, turn documents into text and write alt text, the same work Opensolr does for the pictures and PDFs on your site.
Every call here goes to https://api.opensolr.com. Each picture read counts against the monthly AI allowance of the account that owns the index, and a picture read before costs nothing; the exact cost of each call is on API Quota.
Pictures
image_to_textWhat a picture shows plus the text printed in it, as search words: what powers search by image.
Referenceimage_clipWhat a picture shows, as words, without reading the text in it.
image_indexWords and a meaning vector for several pictures in one call, ready to index.
Referenceimage_ocrThe text printed in pictures: receipts, labels, screenshots.
Referenceimage_alt_textAlt text for a picture, written to your own instructions, in the language you choose.
Documents
doc_to_textThe text of documents sent as files (PDF, Word, RTF, OpenDocument text, HTML, plain text), a PDF page by page with the pictures inside it read.
ReferenceThe same work, done for you while your site is read: Search by image, Text in scans and pictures and Pictures described by what they show.