Skip to content
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

Setting for API key in the UI #1146

Closed
rbren opened this issue Apr 15, 2024 · 4 comments · Fixed by #1319
Closed

Setting for API key in the UI #1146

rbren opened this issue Apr 15, 2024 · 4 comments · Fixed by #1319
Assignees
Labels
enhancement New feature or request severity:medium Problems that affect many users

Comments

@rbren
Copy link
Collaborator

rbren commented Apr 15, 2024

What problem or use case are you trying to solve?
Let the user paste their API key into the UI

Describe the UX of the solution you'd like
A password box in the settings modal.

It should be saved in local storage, and associated with the selected model.

If the user changes to a new model, we should clear the API key input, but persist the old API key in localstorage

Open question: how do we note that some models use the same API key (e.g. gpt-4 and gpt-3.5)? Should we just have one key for now?

Do you have thoughts on the technical implementation?
The API key should be sent via the initialize event, and will need to be used when constructing an LLM object.

@rbren rbren added the enhancement New feature or request label Apr 15, 2024
@li-boxuan
Copy link
Collaborator

I think the GUI should also prevent users from modifying API key when a task is executing.

@rbren
Copy link
Collaborator Author

rbren commented Apr 17, 2024

@li-boxuan that does sound reasonable! Honestly all settings should probably be frozen while a task is in progress

@JustinLin610
Copy link
Contributor

Being frozen is quite necessary. I may prefer that people first finish the setup first, and then run the whole task (now it seems that everything is still a single turn). We may have some reset button, then users can setup the key and model for the next task.

Now there is another pain is that the key is actually highly correlated with the model. Do we need to have some check to make the key and the model work first before entering the task?

@Sparkier Sparkier self-assigned this Apr 23, 2024
@Sparkier
Copy link
Collaborator

@li-boxuan that does sound reasonable! Honestly all settings should probably be frozen while a task is in progress

Implemented in #1291

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request severity:medium Problems that affect many users
Projects
None yet
Development

Successfully merging a pull request may close this issue.

4 participants