it seems this endpoint can only used images in messages array.
api. openai. com/v1/chat/completions
have someone managed to analyze a pdf file with result?
as example “summarize this pdf document”
https://platform.openai.com/docs/api-reference/chat
https://platform.openai.com/docs/api-reference/files
Is the pdf a scanned image? If it is then you need to extract the image and then analyze the image. In Linux I use pdftoppm to extract the image. You can also check out ocrjob for Xojo at GitHub - gregorplop/ocrJob: An ocrmypdf front-end / batch job designer
Or use livetext on macos “as is”…
my current solution is pdf pages to images and text+images for the api request because of this end-point.
there is a free online converter tool:
https://www.adobe.com/de/acrobat/online/pdf-to-jpg.html
it only mentioned text and image, not pdf. simular to playground ui.
