Cursor
The estimated inference cost of a daily user exceeds the subscription price, so the plan is priced below its own token bill at this volume.
- Vendor
- Anysphere
- Category
- Coding
- Tier
- Pro
- Price / month
- $20
- Est. API cost
- $27.98
- Est. markup
- 0.71×
- Price checked
- 11 Aug 2026
- Verification
- verified
What the price actually is
Pro includes a monthly allowance of model requests; use beyond the allowance is billed as usage-based overage on top of the $20. This estimate assumes a user staying inside the included allowance, which is the only comparison the sticker price supports.
The usage this is priced against
A daily user working inside the included allowance: about 250 agent requests a month carrying substantial repository context, plus 8,000 tab completions on a cheap fast model.
Someone using it half as much sees double the multiple. The assumption is the argument — if you disagree with it, the number below is not about you. Change it in the calculator.
The cost math, in full
Estimated monthly inference cost
Mid-tier frontier text model at $3.00 / 1M input tokens · 250 agent requests at ~25k tokens of context and ~2k tokens of edits each
Mid-tier frontier text model at $15.00 / 1M output tokens
Nano text model at $0.10 / 1M input tokens · 8,000 tab completions
Nano text model at $0.40 / 1M output tokens
Rates come from a dated rate card of representative published API prices, not from the vendor. Nobody outside these companies knows what they actually pay; volume discounts and in-house serving both push real costs below these figures, which makes every multiple here a floor rather than a ceiling.
Why this verdict, not the number
A markup below 1× means the arithmetic this site performs stops being the interesting question: at the assumed usage the vendor is not marking up inference, it is absorbing part of it and recovering the difference from lighter users and from usage-based overages. Read the multiple as a statement about the included allowance, not about the company's economics overall — the overage pricing is where heavy users actually pay, and this entry does not model that.
What the price buys besides tokens
- Orchestration
- UX craft
- Scale & infra
Multi-file edits that apply cleanly, an index of the repository kept fresh as you type, retries and diff application that mostly do not corrupt the working tree. The boring hard parts are the product; the model call is the trivial part of an agent request.
What you lose if you leave
Repository indexing you did not have to build, apply-and-review UX for multi-file diffs, and the latency budget that makes tab completion usable rather than annoying.
The honest cheaper path
Aider or Cline with your own API key gets close on the agent loop and costs what the tokens cost. At the assumed volume that is estimated to be more than the subscription, not less.
Recompute it for yourself
Recompute this for your own usage
Scaling assumes your usage has the same shape as ours, just more or less of it. If your mix is different — far more output than input, say — the estimate drifts. It is an estimate either way.
Where every number came from
- pricingCursor pricing
- docsCursor documentation
- rateRate card entry: frontier-mid
- rateRate card entry: frontier-nano
Price recorded 11 Aug 2026 · entry last reviewed 11 Aug 2026. Think something here is wrong? File a correction — we publish them, including the ones that embarrass us.