#Mistral OCR down?

7 messages · Page 1 of 1 (latest)

ripe harbor
#

Seems that for small PDFs (~3 pages) it is no issue, however when running or larger PDFs (~100 pages) it times out

ripe harbor
#

if I look at two different PDFs of approx. 150 pages each, one of them works, while the other one does not work. What could cause this? I can't seem to find anything in the documentation that explains this behaviour.

#

If I take the one that does not work, and download it, and then upload the PDF to mistral, then it works. Hmm

lethal tide
#

Small question : what are your files sizes?
The OCR API can handle up to 1000 pages (should be fine), and 50Mb (it may be problematic).
You can also try to split your pdf doc in subpages.

ripe harbor
#

File size of the one that does not work is 13mb, so the limits there is not the issue

#

and 149 pages

lethal tide
#

Ok! And does it work when splitting in subpages?
Also you are using mistral OCR and not document AI ?