I've noticed that ChatGPT can only read text from PDFs and not images, which leads me to rely heavily on screenshots for sharing graphs and complex formulas. The current limit of 10 images per prompt and 200 per day feels pretty restrictive, especially given that the context window could handle more than that. Because of this limitation, I've been exploring Google AI Studio, which doesn't have an image cap—just a limit on the context window. I'm curious if anyone has found workarounds for these image limits? Also, do Gemini and Grok have similar restrictions?
4 Answers
I switched to Gemini, and it allows for infinite image analysis, which is a game-changer!
You know what's wild? ChatGPT will let you animate all sorts of things. Check out this funny music video I made with it—it features Jesus searching for a great massage spot! Pretty hilarious stuff!
Yeah... definitely a unique take on things.
I found a nifty trick! By using the "print as image" feature with a Windows PDF printer, I create a single multi-page PDF that combines everything into images. This works great for my physics exams with graphs and formulas. Just make sure to use the "print as image" option; otherwise, it'll read the text and ignore the images, which can be so frustrating when you want to include high-res pictures!
Does this method really work well for a single large image? I worry that it compresses the file too much before sending.
Honestly, mine can read text from images decently. It's strange you're hitting such limits. Maybe consider creating multiple accounts for more access? What are you using all those images for—animations or something?
Are you using the paid subscription for Google Gemini?