Zed is a high-performance, multiplayer code editor from the creators of Atom and Tree-sitter, written from scratch in Rust. It is designed for speed and real-time collaboration with both humans and AI — leveraging multiple CPU cores and the GPU (via its GPUI framework) to deliver a responsive editing experience.
Key Features
GPU-accelerated UI — the custom GPUI framework renders the interface at high frame rates, for near-instant response even in large files
Multiplayer collaboration — share your workspace with others in real time via Zed Channels and shared links
Native agentic AI support — bring agents like Claude Agent or Codex, or connect MCP servers to extend the editor’s knowledge
Language Server Protocol (LSP) — full language support with inlay hints, syntax-aware selections, and diagnostics
Multibuffer editing — compose excerpts from across your project for cross-file find-and-replace
First-class Vim bindings — modal editing with text objects and marks
Built-in REPL and debugging — native debugging support via the Debug Adapter Protocol (DAP) across many languages
Extensible — hundreds of extensions for language support, themes, and more
Why Use Zed?
Zed targets developers who want a native, low-latency editor that feels responsive on modern hardware. It is especially well suited to pair programming and AI-assisted workflows, and its open-source core means anyone can build from source or audit the code.
Platform
macOS · Linux
Licence
GPL-3.0-or-later (editor source; GPUI UI framework under Apache-2.0)
Overview Trae (TRAE) is an AI-native code editor from ByteDance that embeds agentic AI capabilities directly into a familiar editor and also offers TraeCode (Trae SOLO), an autonomous AI coding engineer that can plan and ship complete features from a description. It is built on a fork of VS Code, so the editing experience and […]
Cursor is an AI-native code editor built as a fork of Visual Studio Code by Anysphere. Rather than bolting AI onto an editor as an extension, Cursor rebuilds the editor around AI agents, with the model understanding your entire codebase and capable of refactoring many files at once, writing tests, and fixing bugs. Key Features […]