I have been using o1-mini and o1-preview in a chat for a few days now and tonight I decided to try changing the model to gpt-4o so that I could attach an image to the chat, I now cannot change the model back to o1-mini because it doesn’t currently accept images. I know it’s a longshot, but is there anything I can do?
2 Likes
Welcome to the community!
Unfortunately, there’s nothing you can do at this time once you switch to 4o as they’re set-up a bit differently on the backend. This might change in the future, though.
1 Like
To be honest, I have the same idea. You mean that you can’t see the thinking process when using o1-mini or o1-preview. And the error rate is also very high
I have this problem, and it has stopped me from using o1 for a long time.
1 Like
It may be a bug, I hope it can be fixed
Now it’s fine again, intermittently
1 Like
I managed to generate an image and switch back to o1 models by:
- first i’ve used o1 for a longer chat
- switched to 4o model and prompted it to generate an image.
- saved the image to my hard drive
- edited the propt(s) that generated an image to NOT generate an image but something else instead
- now that there were no longer images in the chat, i was able to switch back to the o1 model
2 Likes
@lassemon open a new chat with o1 and say something like “remember this thing that we were chatting about? this was the most recent thing you said about that chat " wkegfawudgsdhjg” let’s continue from there and pick that longer chat up again