Cursor Pro is $20 a month, Pro+ $60 and Ultra $200, and a Teams seat is $40 per user ($32 billed yearly). Past the included usage, on-demand usage continues at API prices, plus $0.25 per million tokens on third-party models for Teams.
Claude CodeFable 5.1 (max) (with fallback)70.4 · Coding Agent Index
CodexGPT-6 Astra (max)67.0 · Coding Agent Index
Muse CodeMuse Spark 1.3 (xhigh)64.2 · Coding Agent Index
Grok BuildGrok 4.5 (high)64.1 · Coding Agent Index
Kimi Code CLIKimi K362.6 · Coding Agent Index
OpencodeGemini 3.8 Flash (high)61.2 · Coding Agent Index
Cursor CLIComposer 2.5 Fast38.3 · Coding Agent Index
Demo setups use current model families. The pairings above are the combinations benchmarked by Artificial Analysis; rankings measure performance, not popularity.
Illustrative daily trends · Other totals: last 30 days. Build Hours: this week, model-estimated activity including overlap; not engineer hours.
↓ Scroll to explore
USED BY ENGINEERING TEAMS AT
How much does Cursor cost per month?
Every plan includes a set amount of model usage; past it, on-demand usage is billed in arrears. Per person or seat:
Cursor plans, per person or seat. Prices verified October 4, 2026 against cursor.com/pricing and Cursor's models and pricing docs.
Plan
Price
What it adds
Hobby
Free
Limited Agent requests
Pro
$20 a month ($16 billed yearly)
Unlimited Tab completions and a set amount of model usage
Pro+
$60 a month ($48 yearly)
More usage; Cursor suggests it for daily Agent users
Ultra
$200 a month ($160 yearly)
Cursor suggests it for power users
Teams Standard
$40 per user a month ($32 billed yearly)
Central billing, usage analytics, SSO and Bugbot code review
Teams Premium
$120 per user a month ($96 yearly)
5x a Standard seat's usage (price as read October 3, 2026; the page no longer shows it)
Enterprise
Custom
Pooled usage, invoicing, SCIM and audit logs
Typical spend
$60–$100 a month for daily Agent users, $200 or more for power users
Cursor's own guide, in total usage
What does the Cursor Token Rate add to a heavy month?
Our own team barely runs Cursor (it co-authored 7 of the 3,192 commits in our two repos from July 5 to October 3, 2026), so we have no Cursor bill to show. We can put the token rate on a heavy month of Claude Code tokens: it applies to cached tokens too, and those are almost all of them. Cursor's agent would not send exactly these tokens.
One engineer's 318 Claude Code sessions (September 4 to October 3, 2026), at the Cursor Token Rate verified October 4, 2026 against cursor.com/docs/models-and-pricing.
Measured
Value
Tokens
67.9 billion, 98.4% of them cache reads
At Anthropic's API list prices
$26,072
Cursor Token Rate on the same tokens
$16,986
Cursor API pricing per token
Past a plan's included usage, Cursor bills each model per token at its API price. On Teams and Enterprise, the token rate comes on top for third-party models, including Auto and your own API keys.
What Cursor charges per million tokens past the included usage, and the Cursor Token Rate Teams adds on third-party models. Prices verified October 4, 2026 against cursor.com/docs/models-and-pricing.
Model
Input
Cache write
Cache read
Output
Claude Opus 5.5
$4
$5
$0.20
$20
Claude Sonnet 5.5
$2
$2.50
$0.20
$10
GPT-5.6 Sol (promotional until November 21)
$4
$5
$0.40
$20
Composer 2.5 (Cursor model)
$0.50
None
$0.20
$2.50
Grok 4.7 (Cursor model)
$2
None
$0.50
$6
Cursor Token Rate, Teams
+$0.25
+$0.25
+$0.25
+$0.25
Cursor usage limits by plan
Cursor publishes no number for any plan's included usage: every plan includes a set amount, and the models you pick decide how fast it runs out.
Each paid plan has two pools that reset every billing cycle: Cursor Models (Grok and Composer), with much more included usage, and Other Models, the third-party models at their API price. On Teams, a seat's usage belongs to that person and doesn't move between members.
When a team member uses up the Other Models pool, Cursor switches them to Cursor Models. Past both, on-demand usage continues at the same API rates, billed in arrears and never slowed or downgraded.
On Teams, on-demand usage is on by default, and admins can set a monthly team-wide spending limit; per-member limits need Enterprise.
What drives a team's Cursor bill
Four things matter most:
Seats or usage. A seat covers its own included usage, and on-demand usage on top is billed per token. A Premium seat costs 3 times a Standard one for 5x the usage. How a Cursor seat compares with Claude Code's and Copilot's is on Claude Code vs Cursor and Cursor vs GitHub Copilot.
The pool. Cursor's own models (Grok and Composer) get more included usage and skip the Cursor Token Rate; third-party models draw on the Other Models pool at their API price.
Auto. Every Auto mode bills at the list price of the model it routes to.
Long sessions. The token rate applies to cached tokens too, and on Claude Opus 5.5 it costs more than the cache read itself.
How to see Cursor cost per engineer
Cursor's own tools first:
Each developer: the usage dashboard and editor settings show included usage per pool.
Admins: on-demand usage is tracked per user in the admin dashboard, and team analytics show agent edits, messages by model and active users, with CSV export.
Admin API: members, daily usage and spending data, plus per-user spend limits on Enterprise.
Zest: the extension reads each engineer's Cursor sessions and models (and, optionally, tokens and cost from Cursor's usage dashboard) and shows them next to their Claude Code and Codex sessions. More in Cursor usage.
TRY IT ON YOUR TEAM
See what your team ships with AI.
Starter is free (3 seats, no credit card). Pro is $49 a month with 5 seats and Max is $149 a month with 10 seats — priced per workspace, not per engineer, with extra seats at $19 a month.
What is the difference between Cursor Teams Standard and Premium?+
Both are Teams seats, and Premium has 5x the usage of Standard for 3 times the price. Members and admins can have either, and Cursor bills per active seat.
What is the Cursor Token Rate?+
A charge per million tokens on third-party model requests on Teams and Enterprise, including Auto and bring-your-own-key requests, on top of the model's own price. It applies to input, output and cached tokens; Cursor's own models are exempt.
Winding / Teams
