Hi everyone,
In older versions of Cursor, there used to be a setting called “Stream responses”, which allowed the AI to show code/results gradually in a streaming way (word-by-word or line-by-line), instead of dumping the whole response at once.
I just installed the latest version of Cursor, but I can’t find this option anymore. Now, when I ask the AI to write code, it only shows the full block after it’s generated, which feels less interactive.
My question is:
-
Was the Stream responses toggle removed in the new version?
-
Is streaming now handled automatically depending on the model/network?
-
Is there any way to enable the old “streaming effect” (typing-like output) so I can see the AI write code gradually again?
Would appreciate any help or clarification from the team or other users.
Thanks in advance!