-
Notifications
You must be signed in to change notification settings - Fork 2
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Unable to change GPT model #32
Comments
Hey @iamdangry If you haven't already, you also need to run I'm realizing now there probably isn't a good indication that you have to do that 😅 The selected model is per client (similar to the dark mode setting in sb). I think we can improve this though and have the default model change automatically if it's the only one in the list like in your screenshot |
Ahhhh that did the trick. Thanks! Odd that changing the image model in the settings updated automatically? Atleast that seemed to be what was happening based off the console output. Is that the expected behaviour? Think this is what confused me and I didn't bother to look in the command palette like a smart user 😂 Edit: |
Thanks @iamdangry, I'm glad that worked! I pushed a new update so if there's only a single text or image model in the settings, it'll always select it as the current model.
What you saw with the image model is the expected behavior, but I guess I messed something up with some backwards compatibility logic I had from before we had a list of |
Updating the SETTINGS yaml with an alternate GPT model other than the default of 3.5 doesn't appear to take effect.
Snippet from SETTINGS:
Console output says it detects a change but the model doesn't change:
![image](https://private-user-images.githubusercontent.com/46516977/346092981-40343d70-e4d0-4b12-bb7c-47a0dc29dced.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.HUgCmRBORTWvSGENhGyit6LNbskY0Zq9C_kN6WmPm5s)
This is further confirmed by the API Usage Dashboard showing my API calls are to
![image](https://private-user-images.githubusercontent.com/46516977/346090753-2345353f-0613-40ae-b0a7-9b47cda0bce5.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.Ofjf1DqT-w0Owz3AzfxBlwF7jRomZtUoJMH6KfEcwus)
GPT-3.5-turbo-0125
instead ofgpt-4o
I can update the image model fine via SETTINGS and the console logs the change correctly.
Of course, if I have done a dumb, please let me know.
The text was updated successfully, but these errors were encountered: