1

I recently experienced hitting the usage cap for GPT-4. At that moment I got a warning saying “Use default model” and clicked it. This changed the model in the thread from GPT-4 to GPT-3.5. Even after the cooling time for the cap limit has finished, the model will not switch back to GPT-4. I would like some help in switching back to GPT4.

Thank you.

3 Likes
2

I fully agree with the comment. By not having an option to switch back to GPT-4, valuable context in previous messages is lost, and contributes to users having to copy-paste previous messages to achieve desired results.

Implementing this measure would help OpenAI reduce GPT-4 consumption.

2 Likes
3

I also agree. Would love to see this implemented if there is not already a way.

4

This feature is fundamental.

I can’t believe I lost all the context for a conversation I’ve been developing for more than a week, what a terrible user experience…

5

Any updates on this feature? I don’t understand why not enable it?

6

I have the same problem after hitting the cap and opting for default, except now i can either use GPT-4 with no plugins or GPT3.5 with plugins, no way to get back to GPT-4 with plugins. Even if i select GPT-4 for the channel when i select plugins it immediately switches that channel to GPT3.5 plus plugins.

7

Hi,
Having the same “struggles”. Please make switching back to GPT-4 a reality. Please baby baby baby, please! Thanks.

1 Like
8

+1

Now I have to pass the context back to GPT-4, which consumes even more GPT-4 powah…

9

Doesn’t GPT4 only use the last two comments or so for the “context” of the conversation regardless? I believe this is why the feature was considered moot.