You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I find myself writing prompts using OpenAI's Playground, then copy-pasting them back into Pezzo, becuase OpenAI's user interface has some advantages. Here are a few ideas I have to achieve parity:
Biggest Requests
Also add a "Run" button that renders the response into a new message appended to the bottom. Having to use the "Test" button, then go into the popup, scroll down to the response, then copy it, is tedious. Also, that response doesn't include newlines. This is the Issue with graphql-codegen type generation when starting the server #1 thing that has me going back to OpenAI's UI.
Add the "Run" button (and the "Test" button) down to the bottom of the UI by the "Add Message" button. If there is a very long chat, it's tedious to have to scroll all the way back to the top to test it.
When clicking the "Add Message" button, have it alternate the speaker. If the last speaker was the user, the next added should be the assistant, and vice versa.
Remove scrollbars from each message, and instead add a scrollbar to the overall conversation. It's so much work to constantly have to resize individual message textboxes to be big enough to read their content. The textboxes would ideally resize to fit all their content, and wouldn't need to be resized or scrolled.
Other Wishlist Requests
Make the font bigger. This is just personal preference, but the font is tiny.
Make the settings panel hideable. Generally we don't edit these settings all that often, compared to editing the messages.
Put the system prompt to the left of the user/assistant messages on wide enough monitors. Generally I'd guess we edit the system prompt and the last message in the chain frequently, and preventing having to scroll a lot in between them is useful.
Is this a feature you are interested in implementing yourself?
No
The text was updated successfully, but these errors were encountered:
I find myself writing prompts using OpenAI's Playground, then copy-pasting them back into Pezzo, becuase OpenAI's user interface has some advantages. Here are a few ideas I have to achieve parity:
Biggest Requests
graphql-codegen
type generation when starting the server #1 thing that has me going back to OpenAI's UI.Other Wishlist Requests
Is this a feature you are interested in implementing yourself?
No
The text was updated successfully, but these errors were encountered: