Extreme overusage

[chadcoop]

Describe the Bug

One request to Sonnet with the word ‘hi’ is using nearly 100k tokens. Average requests are in the hundreds of thousands for extremely small tasks. Have been using Cursor for months, and this was NOT the case just a couple weeks ago when a simple request would typically use a few thousand tokens. Looking at history, the average request is using 10-20x what it was in agent mode just a few weeks ago. No wonder rate limits are hitting so fast.

I am on the Pro+ plan and its now been 2.5 days since I have been able to use sonnet 4. There is no resetting every few hours as described in the plan. And strangely, everytime I make a request and hit the rate limit my usage

This application has become completely unusable.

1. Why are requests suddenly using such an extreme amount of tokens in recent releases?
2. When do rate limits reset on the Pro+ plan?
3. Are there any settings that have been introduced that need to be turned on/off to restore usage behavior to what it was previously?

Steps to Reproduce

Send ‘hi’ as a request and look at dashboard usage.

Operating System

MacOS

Current Cursor Version (Menu → About Cursor → Copy)

Version: 1.2.1 (Universal)
VSCode Version: 1.99.3
Commit: 031e7e0ff1e2eda9c1a0f5df67d44053b059c5d0
Date: 2025-07-03T06:08:06.355Z
Electron: 34.5.1
Chromium: 132.0.6834.210
Node.js: 20.19.0
V8: 13.2.152.41-electron.0
OS: Darwin arm64 24.1.0

Does this stop you from using Cursor

Yes - Cursor is unusable

[condor]

Hi @chadcoop and welcome to Cursor Forum

Thanks for the bug report, to narrow down the possible causes and what is happening:

• Is the 100k tokens on ‘hi’ happening on new chats or those that have previous requests and responses?

If not new chat then its correct as a chat history is also costing tokens though most of them will be “Cache Read” tokens which are the cheapest possible.

have a look on Dashboard > Usage > view: Tokens
It shows there which tokens are being used.

[condor]

Please also have a look at the recent update

Clarifying Cursor Pro Pricing

We want to acknowledge that our recent pricing changes for individual plans were not communicated clearly. That’s our mistake, and we’re sorry. If you were unexpectedly charged for usage between June 16 and July 4, we’ll issue a full refund - just email us at [email protected]. Here’s what’s important to know about the Pro plan: Unlimited usage of Tab and models in Auto $20 of frontier model usage per month at API pricing An option to purchase more frontier model usage at cost We’ve br…

[aldinokemal]

Yeah, for me, just saying “hello” already uses 50k tokens.
This might depend on our MCP setup.
Try disabling all MCPs and check the token usage again.

[wtester]

Pro Plan Rate Limit Transparency Issues - Need Specific Usage Details

At this point, they’ve simply provided a formal response, but not an actual solution to the problem — and they still refuse to acknowledge it. When we make a small request of 50KB and receive a small response of 1–2KB, we’re also charged for some imaginary cache ranging from 500K to 2 million tokens, which inflates the request cost to $1. As customers, we’re paying for either unlimited or a limited number of fast requests, with slower access afterward — not for speed restrictions that last for…