Cursor
The AI code editor and unified agent workspace built on VS Code that turns natural language into working software across your entire codebase.
What it is
Cursor is an AI native code editor developed by Anysphere, built as a fork of VS Code with deep AI integration at every layer. It combines a familiar editor experience with an autonomous coding agent (Composer 2.5, built on the open source Kimi K2.5 checkpoint from Moonshot AI and further trained with reinforcement learning, though Anysphere faced community criticism for not initially crediting Kimi's contribution when Composer 2 launched), a proprietary Tab autocomplete model trained with reinforcement learning, and full codebase understanding through semantic indexing. Cursor supports multiple frontier AI models including Claude Opus 4.8 and Claude 4.6 Sonnet from Anthropic, GPT-5.6 Luna, GPT-5.6 Sol, and GPT-5.6 Terra from OpenAI, Gemini 3.1 Pro and Gemini 3.6 Flash from Google, Grok Build 0.1 from xAI, and its own Composer 2.5 model. With Cursor 3, Anysphere introduced a unified agent workspace built from scratch around agents, featuring multi-repo layouts, seamless handoff between local and cloud agents, and integrations with Slack, Linear, GitHub, Jira, and Microsoft Teams. Trusted by over half of the Fortune 500, it has crossed $2B in annualized revenue and counts millions of developers as users. Anysphere raised $2.3B in its Series D at a $29.3B valuation and is reportedly in talks for a $2B Series E at a $50B valuation.
Who it is for
Developers who want an AI-powered editor that feels like VS Code but with deep, native AI capabilities at every step
Teams building complex applications who need shared rules, analytics, SSO, and centralized configuration
Non-technical founders and product managers exploring code-level control through natural language prompts
Developers who want multi-model flexibility, switching between Claude, GPT, Gemini, and Grok models as needed
Anyone migrating from VS Code who wants AI coding features without learning a new editor or switching to the terminal
Engineering organizations at scale looking for enterprise security, SCIM provisioning, and audit controls
Pros and cons
Pros
- +Built on VS Code, so extensions, themes, keybindings, and muscle memory carry over with zero learning curve.
- +Multi-model support lets you choose the best AI model for each task: Claude for reasoning, GPT for speed, Gemini for large context.
- +Tab autocomplete is exceptionally fast and accurate, trained with reinforcement learning on real-world coding scenarios.
- +Agent mode handles complex multi-file tasks autonomously: planning, coding, running tests, and iterating until done.
- +Trusted by over half the Fortune 500 including Stripe, NVIDIA, OpenAI, Figma, and Salesforce (90% developer adoption at Salesforce).
- +Subagents run in parallel with different models, dramatically accelerating complex tasks that involve multiple parts of the codebase.
- +Rapid innovation pace with major releases every few weeks. Cursor 3 (unified agent workspace), Composer 2.5, automations, Jira integration, and shared canvases all shipped in the first half of 2026.
Cons
- −Primarily tied to the Cursor editor. JetBrains support now exists via ACP, but unlike CLI agents, you still cannot use it inside your preferred terminal or Vim setup.
- −Usage limits on agent requests can be frustrating. Pro users may hit rate limits during extended coding sessions.
- −Pricing adds up quickly at the team level. At $40 per user per month, a 20 person team pays $800 monthly before the Enterprise tier.
- −Codebase indexing on very large monorepos can take significant time on the initial setup, though subsequent queries are fast.
- −As a VS Code fork, it can lag behind upstream VS Code updates and occasionally has compatibility issues with certain extensions.
Pricing
- No credit card required
- Limited Agent requests
- Limited Tab completions
- Extended limits on Agent
- Generous limits for Grok & Composer
- $20 of API usage per month
- Generous Auto + Composer usage
- Everything in Pro
- 3x Pro limits on Agent requests
- $70 of API usage per month
- Everything in Pro
- 20x Pro limits on Agent requests
- $400 of API usage per month
- Everything on Individual plans
- 5x Standard limits on Agent requests
- Access to Grok Bot
- Everything in Teams
- Pooled usage to maximize budget efficiency
- Invoice/PO billing
Prices change; check the official pricing →