I can put in prompts in the Chat mode of Playground and submit them, and they work. However, I can’t save, load, or view code at all.
Do I not have access to the gpt-3.5-turbo API? I ask this because when I try to use the API out side of playground I get “Error with OpenAI API request: openai.ChatCompletion is not a function”
I know there is a waitlist for GPT-4, but I thought gpt-3.5-turbo was generally available in beta.
I believe that the Save, Load and View Code buttons are currently unavailable for the Chat mode because there is no API call that can define the system yet. I’m sure that once the feature becomes available in the API that it will become available in the Playground as well.
How could this be?!?! Don’t other developers already have access to the Save/View Code features in the updated API Playground (March 27, '23)?
Here is the message that I received from OpenAI acting as if this is the case for everyone. I find that extraordinary hard to believe. My guess is that there is a problems with a smaller subset of users that, for some reason, don’t have the ability to use the API.
Can we escalate this?
Here is the message I was sent:
Hello Brian,
Thank you for reaching out to OpenAI Support.
We apologize for the confusion and inconvenience this has caused you. Unfortunately, we don’t have a ‘Save’ feature under Chat (Beta) for now. We appreciate your input and have documented it for internal review.
If you have further comments or suggestions, please do not hesitate to contact us.
This can’t be the case for everyone. Is there any way to use the Chat (Beta) model??? Any workarounds? According to this woman from OpenAI, no one in the world has access to this, but that doesn’t seem to be the case?!?
I agree, this is mindless rubbish. It should be the easiest thing in the world, and yet it seems impossible. It’s like trying to connect to SQLServer from Visual Studio 2022
Same for me, Load/Save/View grayed out and this is a logical feature, after I tried a chat, I would like to continue at a later time at the same point. Maybe OpenAI is surprised by the usage of the playground and thought this is not worth the effort?
As well from me, this feature Load/Save/View would be very much appreciated!
For what it’s worth, I’m having the same problem: in the Chat mode in Playground, the “Load a preset,” “Save,” “View Code,” and “Share” buttons are grayed out and not working.
Which means I can’t view and copy the code for my prompts and settings.
Which apparently means I can’t integrate gpt-4 or gpt-3.5-turbo into the app I’m building, at least right now.
Looking forward to this being enabled on the Chat layout - it would be valuable to have the View Code since it is the more challenging request layout. Came looking here to confirm I wasn’t missing something - I guess we will have to wait a few more weeks