Thanks for reporting a bug you have found in Cursor!
Please add the following info to help us diagnose your issue:
Check the forum to ensure the issue hasn’t been reported already Provide a clear description of the bug Explain how to reproduce the bug (if known) Attach screenshots or recordings (e.g., .jpg, .png, .mp4). Tell us your operating system and your Cursor version
Version: 0.49.4 (Universal)
VSCode Version: 1.96.2
Commit: ec408037b24566b11e6132c58bbe6ad27046eb90
Date: 2025-04-22T00:33:06.906Z
Electron: 34.3.4
Chromium: 132.0.6834.210
Node.js: 20.18.3
V8: 13.2.152.41-electron.0
OS: Darwin arm64 24.5.0 Tell us if the issue stops you from using Cursor.
yes
Right after the update it was fine, everything was working. But after about an hour this story started. I’ve already rebooted my computer and restarted Cursor. I’ve tried new commands, all the same.
It’s disconcerting that all of a sudden indexing has started and it’s still going on. That’s weird
Reinstalled 0.48, any chance there’s a way to restore the 10,000 lines of previous chat history and context from today? I’m not holding out much hope, but just in case?
LOL, I do that too for code and docs obviously, but it’s a shame the chat history doesn’t save to Git alongside that. That’s actually not a bad feature idea!
I dunno, I’ve been noticing a certain widening of the memory function, especially since Claude 3.7. But i agree, overall, memory loss is a hinderance to large codebase progression.
I jump between models all the time. I’ve noticed that after a while, if you work on one model for a long time, it starts to make blatant mistakes. Especially when the cursor writes that 25 requests or whatever are exhausted and continue. At this point, the model loses the thread of the narrative and starts messing around.
Yesterday Claude 3.7 was doing my head in such a way that I was starting to freak out. Then I switched to Gemini 2.5, and in a couple of hours I was able to sort out everything that Claude had done to me.
If you could please share this information:
→ How frequently does this happen? Every time in every chat, on specific chats, every two requests, etc?
→ Are you using fast or slow requests?
→ Screenshot of your console logs when you try a request that fails (Open the command palette, usually with CMD/CTRL + SHIFT + P → Developer: Toggle Developer tools)
I’m having a similar problem where the window will freeze, go into “Not responding” mode and then revert back. Happens mid request and just started today. I’m using slow requests with gemini 2.5 and it’s happening with every prompt. And it usually freezes/unfreezes 2-3 times until I get the response from the model. It’s not usable in its current state.