I got on cursor yesterday afternoon (Saturday) and some models would never respond, others take 5-10 minutes. And it’s still happening today. I’m nowhere near my plan limits or anything.
Yes it is
this happened to me too. In my case it followed cursor just crashing repeatedly. It also updated itself to the newest version after crashing. While it hasn’t asked me to re-login, it continues to freeze/crash.
+1 cursor pro plan, it’s very slow with gemini model.
Same here guys. Also another glitch is, my agents cannot edit code directly in files and try to create duplicate files with tag .new then copy code from old file with new lines and then trying to replace code in original with new code… WTF???
1 Like
Second month in to my pro subscription, slow, poor results, forgets rules and hugely inconsistent. Won’t be renewing.
4 Likes
yes!!! still has problem!!!
The only reason I’m on 0.45 - ffs every next update makes it even worse.
1 Like
Adding to my previous comment. It looks like as a side effect, it is also wasting fast requests. Whether you got fed up with waiting or it somehow “died” generating, it can count the request. I’m out of fast requests now. (not saying all were wasted, just some)
VERY frustrating. I basically only get responses using the “pro” models that are burning $$$ with bugs and incomplete responses. NOT COOL.
I am having to tab over the my browser to have the FREE version of chatGPT handle stuff because Cursor is becoming useless on the “pro” plan.
Funny, I’m actually doing that as well but with Grok.
Im having the same issues i got pro because i thought i hit the fast requests limit however now im burning through the fast requests and waiting for a while for every response.
Side thought (wouldnt it be ironic if this was an issue caused by someone vibe coding?)
Are you hearing this Cursor folks? Do you think it is acceptable for us to be resorting to this - having to leave the Cursor context to have other tools (free tools) pick up the slack?
Look, offer more plans that provide more stability and more context if you have to. Offering only one plan and expecting it to satisfy all your users is completely inadequate.
2 Likes
I don’t know what’s gotten into me lately, but I seem to be in a bad way. Wrote hundreds of lint errors…
I don’t know what changed but the experience is completely different from what it was a month ago. The agent keeps hanging - I’m having to “restore” multiple times. The response is not quicker as it was before, the code quality has gone down drastically. This makes me want to go back on my Pro Plan, especially when Windsurf and VSCode agent are coming up with new updates.
1 Like
Cursor Pro version has been very slow on responses lately. Stuck in “Generating” very often, usually requiring restarting the prompt after more than 10 minutes passed without a response. In Agent (Auto) mode. Sometimes the model responds fast, but usually after 1 or 2 prompts, it is back to being stuck in “Generating”.
Sometimes it times out and asks to “Resume” but usually it does not work either.
Version: 0.49.6
VSCode Version: 1.96.2
Commit: 0781e811de386a0c5bcb07ceb259df8ff8246a50
Date: 2025-04-25T04:44:33.500Z
Electron: 34.3.4
Chromium: 132.0.6834.210
Node.js: 20.18.3
V8: 13.2.152.41-electron.0
OS: Linux x64 5.15.0-138-generic
1 Like
I have a Pro account and most requests are Erroring out. Cursor is seeing this and recording it, but I have quit requesting as half responses are counting against my fast requests, which is unacceptable.
2 Likes
Same here… pro account and unusable and I see ~50% are ‘errored’ and not charged but is charging me for prompts where its “generating” and I have stopped it because its been in that state for >10mins per prompt… the experience is completely useless right now. The model seems to have also forgotten half the project and keeps asking me to point it to relevant code (it wrote only a few prompts back).
Similar issue on my end, ever since o3 and o4-mini were added, both of these models take around 3-5 minutes to provide a response, I dont mind waiting 30 sec for sonnet or gpt 4.1, but o4-mini and o3 are straight unacceptable. I have paid plan. I am seriously considering migrating to a competitor as there are also many bugs, which I have reported, even on this forum and nothing seems to be getting done about that.
1 Like