I just subscribed to the Ultra plan for a year ($2000) after being on the MTM for 6 months ($1200). I was extremely disappointed to see the new percentage based dashboard instead of just showing me the actual costs. It feels very bait and switch. Also update the pricing page in the docs to show exactly how much Auto and Composer we can use instead of just “generous”
“
Each plan includes usage charged at model inference API prices:
Pro includes $20 of API agent usage + generous Auto and Composer usage
Pro Plus includes $70 of API agent usage + generous Auto and Composer usage
Ultra includes $400 of API agent usage + generous Auto and Composer usage
“
EDIT:
Looks like its a result of this call https://cursor.com/api/dashboard/is-on-new-pricing
if you intercept and change to
{ “isOnNewPricing”: false, “hasAutoSpillover”: false }
you can see the old way
does not change the fact that they do NOT state anywhere EXACTLY HOW MUCH the Auto + Comoser bucket allows you.
Edit2:
But based on their stats coming back…
94 million tokens is 2.755% of approximately 3.41 billion tokens.
94,000,000 ÷ 0.02755 ≈ 3,411,978,221 tokens (~3.41B)



