Request ID: e8f1df3e-4ad8-490a-addc-e22c1c9c46c7
ConnectError: [unknown] No response from model
at Lle.submitChatMaybeAbortCurrent (vscode-file://vscode-app/c:/Users/publi/AppData/Local/Programs/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:2068:26077)
Same for me. Been like this for a few days now. Waiting for support to help (going round and round in circles with ‘Sam’ the AI support assistant). I am using a standard consumer fibre broadband account on Windows 11 Pro. No VPN. All has been fine until a few days ago. No changes on my system. It started by saying no internet connection (the internet is ‘always on’ for me - very rarely ever fails). I can also load the standard VSCode and use AI’s with that no problem (and browsers and everything else internet related including writing to this forum right now, while Cursor is telling me I have no internet connection - ‘Sam’ suggested I turn off http/2, restart and try again etc. all of which showed that everything was working fine according to Network Statistics - in all modes http/1 and http/1.1 as well).
Request ID: 64a24e32-d0cf-4101-9296-4c50ed84921f
ConnectError: [unknown] No response from model
at Lle.submitChatMaybeAbortCurrent (vscode-file://vscode-app/c:/Users/jaadb/AppData/Local/Programs/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:2068:22289)
at async Object.Qi [as onSubmit] (vscode-file://vscode-app/c:/Users/jaadb/AppData/Local/Programs/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:4016:16751)
This has now been reported multiple times over multiple weeks without any response from the Cursor team yet, afaik. I’ll respond here since it seems to be the most recent.
I have a potential solution that I found in a different thread on this error. While restarting Cursor, restarting my machine, downgrading Cursor, and restarting my network adapter didn’t fix it, for some reason creating a New Chat did! This may work for y’all too.
Whatever is happening, it’s clear that in many cases it has nothing to do with networking or model availability. Error reporting needs to be improved as well as whatever underlying issue (perhaps a resource limit?) is actually causing the error.