Cursor

An AI code editor where AI edits and writes code directly

Coding

CurrencyUSDKRWEUR

Personal starting price≈€7.13+ ($8)Official price in USD
Free tierFree plan available
Korean supportGood

Price checked 2026-10-17, source: cursor.com/pricing. Prices and benefits may change. Check the official site before you pay.

Pros

  • Understands and edits the whole project
  • Powerful agent features
  • Similar to VS Code, so easy to get used to

Cons

  • Usage-based billing is complicated
  • Costs rise with heavy use

These are opinions compiled by Pickx from public sources.

At a glance

Cursor is a VS Code-based AI code editor that offers everything from code autocompletion (Tab) to agents that edit multiple files on their own, cloud agents and a code review bot (Bugbot). You can choose its own model (Composer) or models from several companies, such as Grok, Claude, GPT and Gemini. The biggest recent change is that, with the SpaceX (xAI) acquisition completed in August 2026, the Grok models were brought in on the "Cursor's own models" side. It is a tool for developers and development teams.

Full model list

Cursor's models fall into two groups. The Cursor Models pool (Composer/Grok) has much more generous included usage, and the Other Models pool (Claude/GPT/Gemini and others) is deducted at each model's API rate as is. Prices are per 1,000,000 tokens, input/output.

ModelTypeFeaturesAvailable plans
AutoCoding (automatic selection)
  • Automatically picks the model that fits the task
  • Can choose between cost, balanced and intelligence modes
  • Charged at the rate of the model actually used
All plans
Composer 2.5Coding
  • Coding-specialized model built by Cursor itself
  • Very cheap at $0.5/$2.5
  • Fast
  • Optimized for agent work
  • Has a Fast version ($3/$15)
  • Certain for Premium and above
  • Whether Free/Plus include it needs checking (Cursor Models pool)
Grok 4.7 / Grok 4.7 500kCoding/reasoning
  • Latest Grok from xAI (SpaceX)
  • Standard version $2/$6
  • 500,000-token long-context version $4/$12
  • Each has a Fast version
  • Generous limits on paid plans
Plus and above (Cursor Models pool)
  • Grok 4.6
  • Grok 4.5
Coding/reasoning
  • Previous Grok versions
  • $2/$6
  • Previous model
  • Plus and above
Claude Fable 5.1Coding/reasoning
  • Anthropic's top model
  • Strong at the hardest problems and large-scale refactoring
  • Expensive at $10/$50
Premium and above (Other Models pool, billed separately at API list price)
Claude Opus 5.5Coding/reasoning
  • Anthropic's flagship
  • Strong at long-running agentic coding
  • $4/$20
Premium and above (Other Models pool, billed separately at API list price)
  • Claude Sonnet 5.5
  • Haiku 5.5
Coding
  • Sonnet: balance of speed and performance ($2/$10)
  • Haiku: very fast and cheap ($0.1/$0.5)
Premium and above (Other Models pool, billed separately at API list price)
GPT-5.6 Sol / Terra / LunaCoding/reasoning
  • OpenAI models
  • Sol ($4/$20): top tier
  • Terra ($2/$12): mid-range
  • Luna ($0.2/$1.2): lightweight/low price
Premium and above (Other Models pool, billed separately at API list price)
Gemini 3.1 Pro / Gemini 3.8 FlashCoding/reasoning
  • Google models
  • Pro ($2/$12): strong at long context and multimodal
  • Flash ($0.75/$3.5): fast and cheap
Premium and above (Other Models pool, billed separately at API list price)
Muse Spark 1.3Coding/text
  • Meta model
  • $1.25/$4.25
  • Added September 2026
Premium and above (Other Models pool, billed separately at API list price)
  • The official docs model table may have more models hidden behind "show more". The above are the main models shown in the official docs.
  • OpenAI released GPT-6 (Sol/Luna/Astra) at the end of September 2026, but the Cursor model docs at the time of checking still show only the GPT-5.6 family (whether GPT-6 is supported needs checking).
  • On Teams/Enterprise, using outside models adds $0.25 per 1,000,000 tokens (excluding Composer/Grok).
  • Max Mode (longer context) is offered only on the old request-count-based plans.

Summary by plan

PlanPrice (monthly billing)What you can doUsage limitsFeatures
Free$0
  • Agent
  • Chat
  • Limited number of uses
  • Exact number not published officially
For a taste
Plus$8
  • Shared AI usage pool
  • Grok-family models only
  • Cloud agents and automations not included
Not published officiallyCheapest entry-level paid plan
Premium$30
  • Latest Grok models
  • Cloud agents
  • Automations
  • Use outside Claude/GPT/Gemini models (billed separately from the subscription at API list price)
4x the usage of PlusStandard plan for individual developers who also use outside models
Super$100Priority model allocation even when usage peaks15x the usage of PlusFor developers who run lots of agents
Ultra$200First to use new products and models50x the usage of PlusFor power users who code with AI all day
TeamsFrom $30 per user (varies by seat tier)
  • Shared usage for the whole team
  • Central billing and management
  • Usage analytics
  • Team-level privacy settings
  • SAML/OIDC SSO
  • Mixed seat types
  • Team marketplace (share rules/skills/plugins)
4-50x Plus depending on seat tierFor development teams
EnterpriseCustom quote
  • All Teams features
  • Invoice payment
  • SCIM
  • Repository/model/MCP access control
  • Audit logs
  • AI code tracking API
  • Dedicated manager
Shared usage at the organization levelFor large companies
  • There is no official Korean won price (the India-only Start plan at ₹649 is not available in Korea). Prices exclude tax.
  • Usage limits refill weekly, and when you run out you can buy extra usage separately. Outside model fees are billed separately from the subscription.
  • As of October 2026, the official pricing page shows no annual billing prices or discounts. The old Hobby, Pro and Pro+ plan names are no longer used.

Which plan for whom

  • Free: Students learning to code, and people who want to try whether Cursor fits their way of working first.
  • Plus ($8): Light individual users who are fine with Grok models alone and want to keep costs to a minimum.
  • Premium ($30): Individual developers and freelancers who code every day. If you need cloud agents and outside models such as Claude/GPT/Gemini, start here.
  • Super ($100): Developers who exceed the Premium limit every month or run lots of agents.
  • Ultra ($200): Solo founders and senior developers who run several agents at once and develop with AI all day.
  • Teams (from $30 per user): Development teams that share rules and skills with teammates and need admin features and SSO. Heavy users can be assigned a higher seat tier.
  • Enterprise: Large company development organizations that need security, access control and audit records.

Notes

Good for

#AI#Artificial Intelligence#Generative AI#Coding#Development#Project Management#Work Management#AI Agent#Free

2 views

Prices converted at 1 USD = ₩1,329.49 (2026-10-19). Original price shown in brackets. Taxes as listed by each provider.

Comments

No comments yet. Be the first.

Your comments and nickname are public. If you see abuse or advertising, report it here and we will review and remove it.

Cursor Plans, models, pros and cons - Pickx