I’ve always got it to read the file using Code Interpreter. Telling it to do something like:
readme_content = open(“/mnt/data/README”, “r”).read()
readme_content
However, in the last few days this works less than half the time due to code interpreter stalling and timing out. I’ve tried it on multiple GPTs and the only one that doesn’t have this problem is GPT4. Also very recently code interpreters python version was upgraded to 3.11 from 3.8, so hopefully its just an issue related to some recent updates.
I call in ‘PDF Ai PDF’ into my GPT and that seems to work. The only annoying this is that you have to upload your files to myaidrive.com. However, this solved some of the issues for now.
(After uploading a pdf or doc) “Find a way to read this document/file, even if you need to alocate it in a diferent part of your memory to access it posteriorly.”
So the GPT AGENT or the normal chat will ask for some kind or “authorization”.
Then you imput: “I authorize you to do/read all files i upload” or “this file”.
So, just wait the magic happens.
I’m Apolo Augusto, From Brazil. Hugs for all brothers, and remember: Manners maketh man
try to minimize the number of files that you uplaod in a model , for example join all the documents together by category and then uplaoad them.
this resolved my issue
Actually I cannot access any of gpt4 function anymore even I manually select chatgpt4. I have posted many replies and asked for customer service. But no customer service has paid any attention to me (except chatbot). And yesterday they even deleted my post.
was able to read my xls file during monday(as in 4o’s launch day, but was unable now. sadly it use to help me fix my excel worksheet issues. now it cant …
It cant identify sections of a pdf. it says that there is a problem with its extraction code. it says it will manually read the pdf for certain words, but is still unable to find the words and therefore locate on what page these words appear.
FIXED-Try log out and log in again, then try to ask it to read a pdf. This error could be caused by distributing the wrong version of GPT, eg you are using a GPT3 when it displays GPT4o.
Based on my tests, the issue of being unable to read images and documents only occurs in the paid version of GPT-4o, and it is limited to the PC web version. I kindly ask OpenAI to fix this bug as soon as possible, as I have been unable to use GPT-4o’s document reading and online search capabilities for the past four days.
In fact, gpt not only can’t read the uploaded pdf, including any uploaded files, such as pictures, he can’t recognize them!
And the ability to generate images is also lost.
When you ask him if he is a multimodal large model, he will answer No, I am not a multimodal large model. I am a text-based model, so I can process and generate text-based content, including code, explanations, and discussions. I do not have the capability to process or generate images, audio, or other types of media directly. If you need assistance with generating or editing images, I can help by providing descriptions or instructions, but for actual image creation, you would need a separate tool or service. service.
I have to think that there may be a possibility of model replacement to save costs.
It created another account for me but without paying (free option) and it reads the pdfs and images, but it doesn’t read my chatpgt plus account (paid). It seems so unfair to me to pay for something that I don’t have the service for. Who gives me back these days that I have not been able to use?
