Back
AI summary
Written by AI from the official notes. Check them for exact details.Mistral AI's OCR API now includes paragraph-level bounding boxes and accepts new page parameter formats.
- New
include_blocksfeature provides detailed paragraph-level bounding boxes. - Structural labels for text elements now included in OCR results.
pagesparameter accepts comma-separated digits and ranges.
Why it matters: Developers and users of the OCR API should care for enhanced data extraction capabilities.
Full release notes2 changes
- Introducing
include_blocksin our OCR API. When set totrue, each page returns ablocksarray with paragraph-level bounding boxes and a structural label (text,title,list,table,image,equation,caption,code,references,aside_text,header,footer,signature) in reading order. Learn more in our OCR documentation. - The
pagesparameter in our OCR API now also accepts a string of comma-separated digits and ranges (e.g."0,1,2","0-5", or"0,2-4") in addition to a list of integers.