Hey, thanks for the report. I need a bit more info to look into this:
Cursor version: go to Menu > About Cursor > Copy, then paste it here
Which model are you using? (Claude, GPT, etc.)
Does this happen in Chat mode or Agent mode?
Next time the generation stops halfway, copy the Request ID (top right menu in the chat > Copy Request ID) and share it here
Also worth trying:
Open settings.json and add "cursor.general.disableHttp2": true, then restart Cursor. This has helped other users with similar generation cut-off issues.
If you’re using a VPN or a corporate network, try without them.