Cursor Pricing (2026): Plans, Usage Limits and How to Spend Less
Cursor costs $0, $8, $30, $100 or $200 a month, but the plan price is only the floor. Here's what each tier includes, how the weekly usage pool works, what happens at the limit and how model choice changes the bill.
Cursor pricing now runs on five tiers: Free at $0, Plus at $8 a month, Premium at $30, Super at $100 and Ultra at $200. Teams charge $30, $100 or $200 per seat, matching the same three tiers. Each paid plan includes a weekly pool of usage for Cursor's own models, and Claude, GPT and Gemini are billed on top at API list price.
Here's the catch: the plan price is only the floor. Cursor sells its plans as part of The X Subscription, so the pool is shared with Grok, and anything outside it is paid from prepaid credit. Two people on the same $30 plan can pay very different totals depending on the model they pick. Output rates alone run from $0.50 to $40 per million tokens.
Read the plan table and the model rates below, then use Cursor pricing the cheap way: stay on the smallest plan and choose the model per task.
Cursor's own advice is Premium for new Agent users, Super for daily users and Ultra for power users running several agents.
What each tier unlocks
Free gives you Agent and Chat with limited usage. Plus adds Cursor's own models (Grok 4.7, Grok 4.6, Grok 4.5 and Composer 2.5) at fixed settings, plus Grok Chat, Voice, Imagine and Build.
Premium is where the coding features start. It adds Cloud Agents, Automations, code hosting on Origin, Bugbot code review and third-party models. Super adds priority model access at peak times. Ultra adds earliest access to new products and models.
Plus is easy to misjudge. At $8 it looks like the cheap coding plan, but it has no Auto, no Cloud Agents and no way to pick Claude or GPT. It suits people who mostly want Cursor's own models and the Grok apps.
How the weekly usage pool works
Each paid plan includes one pool of usage for Cursor models. Agent in the editor, the CLI, Cloud Agents, Grok Bot and the Grok apps all draw from it. The pool refreshes weekly, so one heavy day doesn't use up your month.
Different models draw from the pool at different rates, and your remaining usage and reset date show in your editor settings. The Grok API is separate and doesn't draw from the pool.
The multipliers matter more than the dollar figure. Premium gives four times Plus for $22 more a month. Super gives fifteen times Plus for $92 more, which is a better ratio per dollar, but only if you really use that much.
What happens when you hit the limit
Cursor's documentation lists three options: wait for the weekly reset, use prepaid credit, or move up a tier. Requests are never downgraded in quality or speed.
Prepaid credit comes from top-ups, or Auto Top-up, which refills when your balance falls below a threshold you set. A monthly additional usage limit caps how much extra you can spend, whatever the credit balance. Set that cap before you turn on Auto Top-up, or a long agent run can drain it.
Credit also pays for third-party models. Those cost API list price from the first request, even when your pool is full.
How model choice changes the bill
All rates below are dollars per million tokens, from Cursor's pricing documentation on 14 Nov 2026. The first two rows draw from your pool. The rest are third-party models paid from credit.
Model
Input
Output
Composer 2.5
$0.50
$2.50
Grok 4.7
$2
$6
Claude Haiku 5.5
$0.10
$0.50
Claude Sonnet 5.5
$2
$10
Claude Opus 5.5
$4
$20
Claude Fable 5.2
$8
$40
Take a hypothetical job of 1 million input tokens and 200,000 output tokens, ignoring caching. Composer 2.5 costs $1. Sonnet 5.5 costs $4. Opus 5.5 costs $8. Fable 5.2 costs $16, sixteen times the cheapest Cursor model for the same text.
Fast variants cost more. Grok 4.7 (Fast) is $4 input and $12 output, double the standard rate. Leave Fast off unless you're waiting on it.
Teams and Enterprise
Teams are priced per seat. Each member gets a Premium, Super or Ultra seat at the same price and weekly usage as the matching individual plan, and you can mix tiers on one team. Plus isn't available for teams.
Teams add centralized billing, usage analytics, team-wide privacy mode, custom Chat data retention, SAML/OIDC SSO and a team marketplace for rules, skills and plugins. Enterprise adds pooled usage, invoicing, SCIM and granular spend controls.
There's one extra cost. On Teams and Enterprise, eligible third-party model requests carry a Cursor Token Rate of $0.25 per million tokens on top of API pricing. Individual plans don't pay it. A team that leans on Claude all day should add that to its estimate.
Legacy plans and old advice
Cursor's documentation says Hobby, Start, Pro, Pro+, Ultra and Teams Standard and Premium seats continue for existing customers. They use two monthly pools and on-demand usage instead of top-ups.
That means older Cursor guides can describe a plan you can't buy now. If a blog post quotes a Pro plan or request counts, check it against the live page. If you're on a legacy plan, the page says it keeps working as it is.
Cursor against the other coding tools
If your main job is shipping apps from prompts, a builder can cost less than an editor plus credit. Our list of the cheapest vibe coding tools compares them, our Lovable vs Bolt.new guide weighs the two biggest builders, and current Lovable coupons and Replit coupons are worth a look before you pay for a second tool.
If you only want Claude's models, compare the cost of Claude inside Cursor with Anthropic's own plans in our guide to Claude pricing. Paying Claude's API rate through Cursor credit is the same list price, so the question is whether you value the editor, not the model. Check Claude coupons too.
The five-minute routine to spend less
Start on Plus or Premium, never Super, and look at your usage after one week.
Set the monthly additional usage limit before anything else.
Use Composer 2.5 or Grok for routine edits. Switch to Sonnet or Opus for hard problems only.
Keep Fast mode off.
Move to annual billing only after a few months of steady use. It saves $60 a year on Premium.
Prices and plan details checked on 14 Nov 2026 from [Cursor's pricing page](https://cursor.com/pricing) and [Cursor's pricing documentation](https://cursor.com/docs/account/pricing). Based on public pricing and documentation, not hands-on testing. Prices exclude tax and change often, so check the live pages before you buy.
Our take
Cursor pricing is cheaper to start than it looks and easier to overspend than it looks. The plan price is fixed, but third-party models, top-ups and Auto Top-up are where bills grow. Most people should start on Premium at $30 if they want Claude or GPT inside the editor, and on Plus at $8 only if Cursor's own models are enough.
Don't buy Super or Ultra on day one. The usage multipliers are real, but you can see your pool and reset date in settings, so let a week of actual use decide. When you do outgrow Premium, compare Super with simply topping up credit, because credit is paid only as you use it.
The same rule from our guide to saving money on software subscriptions applies: pay for the usage you have, not the usage you might need. Browse current Cursor coupons or more AI tool deals before you upgrade. Pick the model per task and Cursor pricing stays predictable.
FAQs
You can wait for the weekly reset, use prepaid credit through top-ups, or upgrade to a higher tier. Cursor says requests are never downgraded in quality or speed. A monthly additional usage limit caps how much paid extra usage you can consume.
Topics
ai tools
C
Written by
Coupo9 Editorial
Every statistic in our guides names its source, and prices come from the brand's own pricing page. When the facts behind a guide change, we update it or take it down.
Midjourney Pricing (2026): Plans, GPU Time and Which to Pick
Midjourney costs $10, $30, $60 or $120 a month, but you're really buying GPU hours. Here's what each plan includes, how many images and videos the hours cover, when Relax mode makes Standard the whole bill and why annual billing is riskier than it looks.