What Is Cursor?
Cursor is an AI-first code editor built as a fork of Visual Studio Code. Launched by Anysphere in 2023, it rapidly became the most talked-about vibe codingVibe CodingA style of software development where a developer describes intent in natural language and an AI tool writes the code — shifting the developer's role from typing code to directing an AI.Learn more → tool — achieving 100,000 paying users by early 2025 and a $9.9 billion valuation. The premise is simple but powerful: instead of bolting AI onto an existing editor as an extension, Cursor rebuilds the editing experience around AI from the ground up.
Because it's built on VS Code, Cursor is immediately familiar. Your extensions, keybindings, and settings largely carry over. The difference is in what Cursor adds: a codebase-aware AI that can read, edit, and reason about your entire project — not just the file you have open.
Key Features That Set Cursor Apart
Composer (now called AgentAgentAn LLM-powered system that can take actions, use tools, and pursue multi-step goals autonomously without human input at each step.Learn more → in Cursor 0.45+) is Cursor's flagship feature. It lets you describe a multi-file change in plain English — 'add authentication middleware that validates JWT tokens and protects all /api routes' — and Cursor plans the changes, edits the relevant files, and shows you diffs to review. It can also run terminal commands, read error output, and iterate. This is the feature that made Cursor feel genuinely different from tab completion.
Cursor's codebase indexingCodebase IndexingThe process of analyzing and converting entire code repositories into searchable embeddings or symbol maps that enable AI coding tools to quickly find and retrieve relevant code context.Learn more → is another differentiator. When you open a project, Cursor indexes your code and stores embeddings locally. When you make a request, it retrieves the most relevant context automatically — you don't have to manually copy-paste functions to give the AI context. The @-mention system lets you explicitly reference files, symbols, docs, and even web pages in your prompts.
Models and Pricing
Cursor supports multiple AI backends including Claude 3.5 Sonnet, Claude 3.7 Sonnet, GPTGPTGenerative Pre-trained Transformer — the model architecture and family name behind OpenAI's most famous models, from GPT-2 to GPT-5.Learn more →-4o, and Cursor's own 'cursor-small' model for fast completions. The premium tier (Pro, $20/month) gives you 500 fast requests and unlimited slow requests per month. The Business tier ($40/user/month) adds team features, SSO, and zero data retention. A free tier exists with limited requests.
The model selection matters significantly for quality. Claude 3.7 Sonnet with extended thinking is the best choice for complex multi-file refactors. GPT-4o is faster and better for quick questions. cursor-small handles autocomplete without burning premium request budget. Many power users bring their own API keys to avoid request limits.
Strengths and Weaknesses
Cursor's main strengths are its deep codebase integration, multi-file editing capability, and the quality of its AI features in practice. Users who have switched from GitHub Copilot often report dramatically higher productivity for complex refactors and feature additions. The UI is polished and the experience feels coherent rather than bolted-on.
The weaknesses are the subscription cost on top of existing VS Code plugins, occasional hallucinations in large codebases where the retrieved context isn't quite right, and some extension incompatibilities (though most VS Code extensions work fine). Privacy-conscious developers also note that your code is sent to Cursor's servers for indexing, which is a dealbreaker for some enterprise environments.
Who Should Use Cursor?
Cursor is best suited for developers who spend significant time on complex, multi-file changes — refactors, feature additions, debugging in large codebases. If you're already using VS Code and find yourself frequently jumping between files to give AI the right context, Cursor solves that problem well. The productivity gains are most pronounced on codebases you know well but that are large enough that AI context is a bottleneck.
It's less compelling for developers who primarily write in JetBrains IDEs (though a JetBrains plugin exists), those with strict data-residency requirements, or those who mainly use AI for simple autocomplete where GitHub Copilot or Codeium serve them adequately at lower cost. Try the free tier before committing — many developers find it changes how they work within a week.