How do you configure the Cursor AI usage limit in the Cursor IDE?

Updated 2026-07-15AI-assisted draft · source-checked1,600 searches/moRanked #236 of 534· Cursor AI
Short answer

You set it on the web dashboard, not inside the IDE. Go to cursor.com/dashboard/spending, open the Spending tab, and set a monthly spend limit — on-demand usage must be enabled for the limit to appear. Changes apply immediately. To spend nothing extra at all, just disable on-demand usage. As of 2026-07.

Why — the first-principles explanation

The reason this is confusing — and the reason people search for it — is that "usage limit" means two different things in Cursor, and they live in different places.

Your included budget is fixed by your plan and you can't configure it. Every plan comes with a monthly API usage budget charged at model provider prices with no markup: Pro ($20/mo) includes $20 of API agent usage, Pro+ ($60/mo) includes $70, and Ultra ($200/mo) includes $400. That budget resets on your subscription anniversary and does not roll over. Separately, there's a First-party models pool for Auto, Composer 2.5, and Grok 4.5 that Cursor describes only as "generous" — it's tracked apart from the dollar budget. Nothing in the IDE lets you raise or lower any of this.

The thing you actually configure is the ceiling on overage, and this is where the money risk lives. When your included budget runs out, one of two things happens depending on a single toggle. If on-demand usage is off, requests stop and you pay nothing more. If it's on, you keep working and get billed at API rates on a separate invoice. A spend limit is the brake on that second path — a hard monthly cap on on-demand charges. Here's the catch that trips people up: on-demand usage must be enabled before the spend limit control is even visible. If you don't see the setting, that's usually why.

So there are really two safety strategies, and you should pick deliberately. Turning on-demand off is the absolute guarantee — you can never be charged more than your subscription, but your AI features simply stop mid-month. Turning it on with a spend limit buys you headroom up to a number you choose. Note what a spend limit does when hit: AI features stop for that user until the next billing cycle. It's a cap, not a warning. Spend alerts are the separate feature that emails you at a threshold — and alerts do not stop usage. Confusing those two is how people get a surprise invoice.

One asymmetry worth knowing on Enterprise: when a user has limits from multiple sources — a member override, a group override, and a team default — Cursor applies the highest applicable limit, not the lowest. And changing the team limit does not overwrite existing member overrides; those persist until you clear them.

An example that makes it click

Think of your plan as a prepaid phone card and on-demand usage as the roaming switch on your phone.

The card is worth $20. When it runs out, roaming-off means your phone just stops making calls — annoying, but you'll never get a bill. Roaming-on means calls keep connecting and the meter starts running. The spend limit is you telling the carrier, "cut me off at $50 of roaming, no matter what." And a spend alert is a text message saying "you've hit $50" — which is a lovely thing to receive while the meter keeps right on running. One of those stops the money. The other just tells you about it.

How to do it

  1. Decide first: do you want a hard stop at your subscription price, or paid headroom? For a hard stop, skip to step 6.
  2. Go to cursor.com/dashboard and sign in with the same account you use in the IDE.
  3. Open the Spending tab (cursor.com/dashboard/spending). Note: the spend limit control is only visible if on-demand usage is enabled — enable it first if you don't see the option.
  4. Set your monthly spend limit for yourself or your team. Changes take effect immediately; setting it to 'No Limit' removes the cap entirely.
  5. Optionally add a spend alert (cursor.com/dashboard) to get an email at a threshold you choose — but remember alerts only notify, they do not stop spending.
  6. To guarantee zero extra charges instead: disable on-demand usage in your dashboard settings. Requests then stop once your included usage is consumed.
  7. Check where you stand any time at cursor.com/dashboard under Usage, which shows real-time usage, remaining allowance, and any on-demand charges; Billing & Invoices splits Included Usage from On-Demand Usage.
  8. Enterprise admins only: set member limits via the Members tab, group limits via the Groups tab, or a team default via the Spending tab — and remember the highest applicable limit wins.

Key facts

Infographic: How do you configure the Cursor AI usage limit in the Cursor IDE — short answer and key facts
Visual summary — How do you configure the Cursor AI usage limit in the Cursor IDE?
CA
See plans for Cursor AI by Anysphere

Open the current plans, credits, usage limits and cancellation terms before you subscribe.

See plans & limits ↗
▶ The 60-second explainer (script)

How do you set a usage limit in Cursor? First, the thing that trips everyone up: it's not in the IDE. It's on the web dashboard. Go to cursor dot com slash dashboard, open the Spending tab, and set a monthly spend limit. Changes apply immediately. But here's what you actually need to understand. There are two different limits. Your included budget is set by your plan and you can't change it — Pro is twenty dollars a month and includes twenty dollars of API usage, Pro Plus is sixty and includes seventy, Ultra is two hundred and includes four hundred. That resets monthly and doesn't roll over. What you configure is the ceiling on overage. When your budget runs out, one toggle decides everything. If on-demand usage is off, requests just stop and you pay nothing extra. If it's on, you keep working and get billed at API rates. The spend limit is the brake on that second path. And one catch: on-demand has to be enabled before the spend limit control even shows up. If you can't find the setting, that's why. Last thing — don't confuse spend limits with spend alerts. A limit stops the spending. An alert just emails you while the meter keeps running.

What authoritative sources say

Cursor Docs — Spend limits (official)official — Spend limits are set in the dashboard's Spending tab, require on-demand usage to be enabled, take effect immediately, and when reached stop AI features for that user until the next billing cycle; Enterprise applies the highest applicable limit across member, group, and team sources. source ↗
Cursor Docs — Usage and limits (official)official — Included usage by plan (Pro $20/mo includes $20 of API agent usage, Pro Plus $60/mo includes $70, Ultra $200/mo includes $400), usage resets monthly on the subscription anniversary, unused usage does not roll over, and the dashboard Usage section shows real-time usage and on-demand charges. source ↗
Cursor Docs — Usage-based charges (official)official — On-demand usage must be explicitly enabled, is billed at API rates with no markup on separate invoices, and can be prevented by disabling on-demand usage or setting a spend limit. source ↗
Cursor Docs — Pricing and plans (official)official — Full plan and price list as of 2026-07: Hobby free with limited usage, Pro $20/mo, Pro+ $60/mo, Ultra $200/mo, Teams Standard $40/user/mo, Teams Premium $120/user/mo. source ↗

People also ask

Why can't I find the spend limit setting?

Almost always because on-demand usage is disabled. Cursor only shows the spend limit control when on-demand usage is enabled — with it off, there's no overage to cap, so there's nothing to configure.

Can I set the usage limit inside the Cursor IDE?

No. Billing controls live on the web dashboard at cursor.com/dashboard. The IDE only shows you a notification when you hit your limit.

What's the difference between a spend limit and a spend alert?

A spend limit is a hard cap that stops AI features when reached. A spend alert just emails you at a threshold and does not stop anything. If your goal is not getting billed, you need the limit.

What happens the moment I hit my spend limit?

AI features stop working for you until the start of your next billing cycle. On a team, only the user who hit the limit is affected — everyone else keeps working.

How do I make sure I'm never charged more than $20?

Disable on-demand usage in your dashboard settings. Then requests simply stop once your included usage runs out, and you pay only your subscription. Note that unused budget does not roll over to next month.

Related questions