Card #483 · Cursor (AI IDE)
Excessive and non-transparent token consumption in newer models
We are experiencing a dramatic and unexplained increase in token usage, particularly with the 5.5 and Composer 2.5 versions. The consumption feels abnormal and non-transparent, often exhausting monthly budgets or daily limits in a fraction of the usual time.
Pain score
0.26/ 1.00 · weighted product of four components
Reach0.70distinct authors (log-normalized)
Recency1.00freshness decay, half-life months
Engagement1.00upvotes + comments, normalized
Monetization1.00willingness-to-pay cues in evidence
§ Evidence — 6 verified quotes
I’ve noticed a dramatic increase in my token consumption.
Our token usage has gone up 5 fold with 5.5.
I consume all my token in 15 minutes with 1-2 task using my chatGPT authentification.
Codex 5.5 is consuming tokens as if its drinking gasoline even at thinking medium
the change from 5.2 to 5.5 is dramatic.
My dashboard suggests that when using codex I used 95 Million tokens input, 25 Million cached input to produce 600K output tokens!!??
Comments are a Pro feature — join the conversation, share what you've shipped, hear what others are building.
Upgrade to Pro