API usage increases when Cursor is idle

Support Bug Reports
1

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

Recently I’ve noticed that Cursor seems to use 2x more API usage quota than, say, a month ago – even though my workflow hasn’t changed. After purchasing “Pro” plan, several quick questions in Ask mode – and already 5% of API usage.

What’s more, API usage increases when Cursor is completely idle and no tasks are performed. Attached is a video with example of such behavior. After request in Chat was finished, 5% of API quota was used. I switched back and forth to other programs; Cursor was not in use. Then, several minutes later (see 02:35 – on video it can be seen that all operations in Cursor are finished) API usage grew to 7% for no apparent reason.

Steps to Reproduce

  1. Make a request in Chat mode
  2. Finish operation
  3. Wait for a few minutes

Screenshots / Screen Recordings

Operating System

MacOS

Version Information

Version: 2.5.25
VSCode Version: 1.105.1
Commit: 7150844152b426ed50d2b68dd6b33b5c5beb73c0
Date: 2026-02-24T07:17:49.417Z
Build Type: Stable
Release Track: Default
Electron: 39.4.0
Chromium: 142.0.7444.265
Node.js: 22.22.0
V8: 14.2.231.22-electron.0
OS: Darwin arm64 24.6.0

For AI issues: which model did you use?

GPT-5.2

Does this stop you from using Cursor

No - Cursor works, but with this issue

3

I think that usage % in the IDE takes time to update. Mine does not reflect instantly after a prompt.

4

Hey, thanks for the detailed report and the video.

The usage percentage in the IDE doesn’t update in real time. There’s a known delay. So what you’re seeing at 02:35, the jump from 5% to 7%, is most likely the delayed display of tokens that were already consumed during your earlier Chat request, not new usage happening while Cursor is idle.

A couple of things to check:

  • Go to your Dashboard and open the Usage tab. It shows a breakdown of individual requests and token counts. This will confirm whether any requests were logged during the time Cursor was supposedly idle.
  • Keep in mind that GPT-5.2 is a premium model and burns through usage faster than smaller models. The “2x more usage than a month ago” could be explained by pricing differences if you switched models recently.

If the dashboard shows requests you didn’t make, that would be a real issue. Let me know what you see there and we can dig deeper.