Free tool · runs in your browser · no signup

The price of a request is decided after you send it.

Since 24 August 2026, every Cursor Auto request bills at whatever model it got routed to. The cheapest and priciest routed models are up to 20x apart for the same tokens, and a request routed to the priciest model runs about 8x what it would have cost under the old flat rate. Slide your real usage in and watch what the routing does to your month. Nothing leaves this page.

These sliders start on illustrative values, not a claim about a typical user. Edit them to match your Cursor usage dashboard, which is the only source that knows your real numbers.

40

Auto / agent requests only. Tab completions are not billed as routed Auto requests, leave them out. Your Cursor usage dashboard has the real count.

80% / 20%

heavy = refactors, debugging, agent runs. We assume Auto sends heavy work 60/40 to Opus 5 / Fable 5, and routine work to Composer 2.5.

Average request size

~2,000 tokens in / 500 out per request

22
The verdict

Estimated routing premium for your assumed mix

$5.15/mo

more than pinning Composer 2.5, at 880 requests a month with 20% heavy work.

Theoretical maximum, if every request went to Fable 5: $37.6/mo. Where you actually land between $0 and that is a routing question your dashboard answers.

Honestly? At your volume the routing barely matters. Go build.

If Auto routed everything to…
20x model-to-model spread
Composer 2.5$0.5 / $2.5 per M$1.98/mo
Opus 5$5 / $25 per M$19.8/mo
Fable 5$10 / $50 per M$39.6/mo

same requests. same you. three price tags.

Your month, itemized
Auto, with your task mix
$0.0081 per request, blended
$7.13
Pinned to Composer 2.5
the whole month at the cheap rate
$1.98
Routing premium (your mix)
72% of your Auto bill is the routing decision
$5.15

estimates at list price, client-side only. your usage dashboard is the ground truth. prices verified against Cursor’s pricing docs on 2026-09-03.

So now what

You cannot control the prices. You can control the routing.

The headline is an estimate for the mix you assumed, and the theoretical maximum is the all-Fable ceiling above it. Which one you are near depends on where the router has actually been sending your requests, and that is knowable: twenty minutes in your usage dashboard settles it. The full walkthrough, including the two cases where switching actually pays, lives in our explainer: What does Cursor Auto actually cost per request now?

And if you want the routing decision back entirely: RelayPlane routes each task to the model you would have picked, with a local spend ledger. Free, open source, MIT.

FAQ

Cursor Auto cost, straight answers.

It multiplies your Auto/agent requests per month by the per-request token cost at each routed model, using the public list prices: Composer 2.5 at $0.50/$2.50 per million input/output tokens, Opus 5 at $5/$25, and Fable 5 at $10/$50. Your task mix decides how much of your traffic we assume Auto routes to the pricier models. The inputs load with illustrative starting values, edit them to match your own Cursor usage dashboard. Everything runs in your browser, nothing is sent anywhere.
Both numbers are real, but they measure different things. Between the cheapest and priciest routed model the spread is up to 20x for the same tokens: Composer 2.5 is $0.50/$2.50 per million and Fable 5 is $10/$50, exactly twenty times more. Measured against history, a request routed to Fable 5 costs about 8x what the same request cost under the old flat Auto rate ($1.25/$6 per million) that Cursor removed on 24 August 2026. So it is 20x model-to-model, and roughly 8x versus the old flat rate, not the same figure.
These are the per-million list prices checked against Cursor's pricing docs on 2026-09-03: Composer 2.5 $0.50 input / $2.50 output, Opus 5 $5 / $25, Fable 5 $10 / $50. Model names follow Cursor's current routed lineup; if Cursor renames or reprices a routed model, your own usage dashboard is the ground truth. Source: https://cursor.com/docs/models-and-pricing.
Two moves. Read your real bill once: open the Cursor usage dashboard, note which model Auto routed each request to, and multiply the tokens by that model’s list price. Then take the routing decision back: pin a cheap model for routine work, or route requests yourself with a tool like RelayPlane, which is free and open source.