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 recently used it to test some backend that handles a responses from LLMs and noticed that Transfer-Encoding: chunked is not supported at all - the client waits for the response to finish to show it on the screen.
It would be a really great enhancement for kulala to support visualizing of each chunk received from API as soon as it comes to the client.
The text was updated successfully, but these errors were encountered:
How should we visualise the chunks when they arrive:
Add notification through vim.notify and update the buffer with response?
How show we deal with the float window? We can close the window, but keep the buffer open and update it in the background. It can the be invoked on keymap or on next chunk arrival.
IMHO, it the view should update the headers according to the last reqeust made (so just replace them maybe?) and the contents of the response chunk should be just appended to the current buffer view.
But maybe @tomaszkiewicz can chime in on that too, to give his 5 cents on how he would like to visualize that?
Let's start with the fact, that I <3 kulala :)
I recently used it to test some backend that handles a responses from LLMs and noticed that
Transfer-Encoding: chunked
is not supported at all - the client waits for the response to finish to show it on the screen.It would be a really great enhancement for kulala to support visualizing of each chunk received from API as soon as it comes to the client.
The text was updated successfully, but these errors were encountered: