Cursor crashes regularly on Mac ever since 2 updates ago

[krisamerikos]

Where does the bug appear (feature/product)?

Somewhere else…

Describe the Bug

While agent is performing tasks, cursor will crash and give the option to restart or close Cursor. When you choose restart, some of the work that was previously done has not been saved. The crashes happen regularly - between 1 and 6 times per hour and it has been like this since 2 updates ago.

Steps to Reproduce

Run the agent on an iMac.

Expected Behavior

The agent should continue working without crashing.

Operating System

MacOS

Current Cursor Version (Menu → About Cursor → Copy)

Version: 2.2.43
VSCode Version: 1.105.1
Commit: 32cfbe848b35d9eb320980195985450f244b3030
Date: 2025-12-19T06:06:44.644Z (2 days ago)
Electron: 37.7.0
Chromium: 138.0.7204.251
Node.js: 22.20.0
V8: 13.8.258.32-electron.0
OS: Darwin arm64 24.6.0

For AI issues: which model did you use?

Auto

For AI issues: add Request ID with privacy disabled

d4dc9e0d-a538-412d-b406-04b49423e761

Additional Information

none

Does this stop you from using Cursor

Sometimes - I can sometimes use Cursor

after reopening, the first new agent prompt shows its generating but never does anything (tested this more than 25 times), but if you stop that prompt and do a new prompt it starts working again… until the next crash

[deanrie]

Hey, thanks for the report.

This is a known issue in version 2.2.x. The team is actively working on fixes for memory leaks and stability.

Please try these temporary workarounds:

1. Start Cursor without extensions:

cursor --disable-extensions

2. If the crashes stop, look for the problematic extension via Cmd+Shift+P > Developer: Open Extension Monitor.

3. If that doesn’t help, disable the GPU:

cursor --disable-gpu

To share with the engineering team, we’ll need logs:

• Cmd+Shift+P > Developer: Open Logs Folder, then attach the log file
• Cmd+Shift+P > Developer: Toggle Developer Tools > Console, then share a screenshot of any errors at the time of the crash

Let me know if any of the workarounds helped.

[David_Weisinger]

Same for me..

[Jon_Decker]

Same here - crashing 30-40 times per day non-stop, losing part of the most recent conversation each time, and the first conversation request after it restart fails every single time. This is bad to the point that customers should be getting partial refunds. How does something this bad get past QA???