Why can't I get Gemini to recognize "strikethrough" text in an image

Yeah you got it. that was javascript. you’d send each screen with that prompt, and then post process the data.

eg in py

1 Like