Cursor Pro
Two monthly pools: Cursor Models and Other Models
Published terms checked 2026-09-29
Works with
Cursor editor · Cursor Agent · Cloud Agents · Cursor CLI
Model access
13 named models
Billing
$20 / month
Two monthly pools: Cursor Models and Other Models
Cursor Models usage rates
USD per million tokens. Fast and context variants are separate. A dash is not a zero rate. These plan-pool rates do not establish a direct API route.
- Grok 4.7
- Input / 1M$2Cache write / 1M-Cache read / 1M$0.5Output / 1M$6
- Grok 4.7 (Fast)
- Input / 1M$4Cache write / 1M-Cache read / 1M$1Output / 1M$12
- Grok 4.7 500k
- Input / 1M$4Cache write / 1M-Cache read / 1M$1Output / 1M$12
- Grok 4.7 500k (Fast)
- Input / 1M$6Cache write / 1M-Cache read / 1M$1.5Output / 1M$18
- Grok 4.6
- Input / 1M$2Cache write / 1M-Cache read / 1M$0.5Output / 1M$6
- Grok 4.6 (Fast)
- Input / 1M$4Cache write / 1M-Cache read / 1M$1Output / 1M$12
- Grok 4.5
- Input / 1M$2Cache write / 1M-Cache read / 1M$0.5Output / 1M$6
- Grok 4.5 (Fast)
- Input / 1M$4Cache write / 1M-Cache read / 1M$1Output / 1M$18
| Model | Input / 1M | Cache write / 1M | Cache read / 1M | Output / 1M |
|---|---|---|---|---|
| Grok 4.7 | $2 | - | $0.5 | $6 |
| Grok 4.7 (Fast) | $4 | - | $1 | $12 |
| Grok 4.7 500k | $4 | - | $1 | $12 |
| Grok 4.7 500k (Fast) | $6 | - | $1.5 | $18 |
| Grok 4.6 | $2 | - | $0.5 | $6 |
| Grok 4.6 (Fast) | $4 | - | $1 | $12 |
| Grok 4.5 | $2 | - | $0.5 | $6 |
| Grok 4.5 (Fast) | $4 | - | $1 | $18 |
Other Models usage rates
USD per million tokens. Fast and context variants are separate. A dash is not a zero rate. These plan-pool rates do not establish a direct API route.
- Claude Fable 5.1
- Input / 1M$10Cache write / 1M$12.5Cache read / 1M$0.25Output / 1M$50
- Claude Opus 5.5
- Input / 1M$4Cache write / 1M$5Cache read / 1M$0.2Output / 1M$20
- Claude Sonnet 5.5
- Input / 1M$2Cache write / 1M$2.5Cache read / 1M$0.2Output / 1M$10
- Gemini 3.1 Pro
- Input / 1M$2Cache write / 1M-Cache read / 1M$0.2Output / 1M$12
- Gemini 3.8 Flash
- Input / 1M$0.75Cache write / 1M-Cache read / 1M$0.075Output / 1M$3.5
- GPT-5.6 Luna
- Input / 1M$0.2Cache write / 1M$0.25Cache read / 1M$0.02Output / 1M$1.2
- GPT-5.6 Sol
- Input / 1M$4Cache write / 1M$5Cache read / 1M$0.4Output / 1M$20
- GPT-5.6 Terra
- Input / 1M$2Cache write / 1M$2.5Cache read / 1M$0.2Output / 1M$12
| Model | Input / 1M | Cache write / 1M | Cache read / 1M | Output / 1M |
|---|---|---|---|---|
| Claude Fable 5.1 | $10 | $12.5 | $0.25 | $50 |
| Claude Opus 5.5 | $4 | $5 | $0.2 | $20 |
| Claude Sonnet 5.5 | $2 | $2.5 | $0.2 | $10 |
| Gemini 3.1 Pro | $2 | - | $0.2 | $12 |
| Gemini 3.8 Flash | $0.75 | - | $0.075 | $3.5 |
| GPT-5.6 Luna | $0.2 | $0.25 | $0.02 | $1.2 |
| GPT-5.6 Sol | $4 | $5 | $0.4 | $20 |
| GPT-5.6 Terra | $2 | $2.5 | $0.2 | $12 |
- Included usage
- Two monthly pools: Cursor Models and Other Models. The current public plan table does not publish an exact dollar or token size for these pools. Source ↗
- Usage pools
- Grok and Composer use the Cursor Models pool; third-party models use Other Models at model API rates. Both reset with the monthly billing cycle. Source ↗
- After included usage
- Paid plans can enable on-demand usage at published model rates, billed monthly in arrears, or upgrade. Source ↗
- Tools
- Cursor editor and coding agent; paid plans include unlimited Tab completions and access to Cloud Agents. Bugbot is usage-billed. Source ↗
- Privacy
- Privacy Mode prevents customer data from being used to train Cursor models. Other Cursor surfaces and model providers have their own data-handling terms. Source ↗
Published terms checked 2026-09-29. Provider terms can change.
Paid plans include both model pools. Choosing a model and its context or speed mode changes usage.
Cursor Models pool4
Provider lineup ↗Fast and extended-context modes have separate usage rates.
Other Models pool9
Provider lineup ↗Third-party model usage draws from the Other Models pool.
Model access checked 2026-09-28. Named access does not establish how much of that model a plan includes or a predictable usage allowance.
Optional on-demand usage at model rates, billed in arrears.
Published terms, sources & history
Published price and access do not establish a numeric allowance.
Included usage
Included Cursor-model and third-party-model usage pools, unlimited Tab, Bugbot and Cloud Agents. Optional on-demand billing.
Grok 4.7 route pricing depends on speed and context
Cursor documents standard and Fast rates plus a long-context tier. Fast is the paid-plan default, but this catalog cannot establish a specific speed route for a recorded workload, so it does not assign one token rate.
Cursor Models pool (included usage, unquantified)
Cursor Models : Significantly more included usage for Grok 4.7, Grok 4.6, Grok 4.5, and Composer 2.5.
Other Models pool (third-party models at API price)
Other Models : The pool for third-party models, charged at the model's API price. Pro, Pro Plus, and Ultra include this pool, with the option to pay for additional usage as needed.
What happens when included monthly usage is exceeded
When you exceed your included monthly usage, you can either: - Add on-demand usage : Continue at the same API rates with pay-as-you-go billing - Upgrade your plan : Move to a higher tier for more included usage. On-demand usage is billed monthly at the same rates. Requests are never downgraded in quality or speed.
Agent limits (extended vs Hobby, unquantified)
Everything in Hobby, plus: Extended limits on Agent / Generous limits for Grok
Tab completions (unlimited)
Pro, Pro Plus, and Ultra include unlimited tab completions, extended agent usage limits on all models, access to Bugbot, and access to Cloud Agents.
On-demand usage toggle (usage-based pricing)
Every plan includes a set amount of model usage. On-demand usage allows you to continue using models after your included amount is consumed, billed in arrears.
What the provider does not publish
Cursor does not publish the size of either usage pool in dollars or tokens for Pro; the docs say only 'Significantly more included usage' for the Cursor Models pool and that the Other Models pool is 'charged at the model's API price'. The docs' 'How much usage do I need?' section gives indicative spend ranges ('Daily Agent users : Typically $60-$100/mo total usage'), which are guidance, not limits.
Model availability scope
Cursor says paid individual plans unlock all supported named models, subject to regional and organization controls.
- Cursor paid-plan pools and current model prices checked Sep 23, 2026
- Cursor pricing (official) checked Sep 23, 2026
- Cursor pricing (official) checked Sep 23, 2026
Catalog record cursor-pro@2026-09-21. Catalog dates identify recorded rule versions. New admissions do not establish a provider launch date. Provider credits are specific to that provider. Listing a price does not establish capacity, equivalent experience or that a plan could replace another.