GPT-4 refuses to analyze pictures and pdf

I subscribed to Plus so that I can use its picture input capability for it to look at my website and give me feedback, which GPT-4 refuses to do.|

he also cant read documents pdf

It says " I’m sorry, but as an AI language model, I am unable to access external links or view PDF documents. However, if you can provide me with a brief overview or key points from the document, I would be more than happy to help you with your request or answer any questions you may have about the content."

1 Like

I believe they’re rolling it out selectively, and maybe through the API first. I’m also waiting. Don’t worry, it’ll be here soon enough! :grinning:

Boa tarde, meu caro.

Não encontrei informações da OpenAI sobre a possibilidade de o Chat GPT Plus realizar as tarefas que você mencionou em seu tópico. Sem informações não posso afirmar ou supor que haverão atualizações que possibilitarão aos usuários solicitar ao CHAT GPT Plus que façam análises, leituras e afins de informações externas ao banco de dados da OpenAI.

If you are look For solution which can chat with pdf. You can try

This also uses openai APIs. And gives accurate results.


I subscribed the GPT plus for its code interpreter function as well, but it refuse to read my
pdf files either. OPENAI mislead me to subscribe but does not offer what they promised.

But the GPT4 did the job nicely after I converted the PDF file into a plain word file and then upload to it.

I apologize for any confusion. The provided output was based on the initial section of the document which was available to me. To provide a more detailed analysis, I can read the entire document and process its contents accordingly. Let’s proceed with the complete analysis now.

Finished working
Show work
Based on the full text, the paper presents a detailed analysis of how cells decide to proliferate or exit the cell cycle. The authors have conducted experimental studies and used mathematical modeling to challenge the traditional understanding of cell cycle commitment, which they argue is not a singular, irreversible event.

Here’s a more detailed analysis:


While the authors provide substantial evidence to support their theory, it is important to consider that their model is based on specific cell lines under laboratory conditions. The behavior of cells may differ in an in vivo environment, or in different cell types. Hence, generalizing these findings to all cells requires further testing and validation.

The authors mention using high-throughput single-cell imaging and tracking, but more details about the experimental conditions, controls used, and replicates would be required to evaluate the robustness of their data.

Don’t get your hopes up. The interpreter uses PyPDF to read PDFs, which means it’s not the best. You’re better off just copying and pasting the unstructured text into it. At the very least if you are using Python I’d recommend using PyMuPDF 1.22.5 documentation instead

If you want a powerful PDF reader you should not use these bargain bin websites. Google has a Document AI that can handle PDFs that even have the text baked into it.

1 Like