Best open-source alternatives to Cursor IDE

Zed is a high-performance code editor aimed at developers who want a fast, collaborative editing experience. It comes from the creators of Atom and Tree-sitter, which suggests a focus on modern editing infrastructure and language-aware tooling. The README presents Zed as a desktop application available for macOS, Linux, and Windows, with direct downloads and package-manager-based installation options. It also points contributors to platform-specific build instructions, indicating that the project is actively developed from source by the community and Zed Industries. The editor is positioned as a multiplayer product, but the README does not detail collaboration features beyond that description.

Codex CLI is a locally running coding agent from OpenAI designed to assist with software development tasks on a user's computer. The README positions it as part of a broader Codex ecosystem, with related experiences available in IDEs and as a desktop app, while also pointing users to a separate cloud-based Codex Web product. The project is aimed at developers who want an agentic coding assistant that can be installed on macOS, Linux, or Windows. Installation is supported through shell scripts, npm, Homebrew, and downloadable release binaries. For authentication, it works either through a ChatGPT account or, with extra setup, an API key.

code-server is a browser-based version of VS Code that lets developers work from virtually any machine while running the editor and intensive tasks on a remote Linux server. It is aimed at people who want a consistent development environment, better performance for builds and downloads, and reduced battery use on laptops while traveling or working remotely. The project is distributed with an install script, manual installation instructions, and guides for deploying it to cloud providers or team infrastructure. The README also points users to setup, configuration, FAQ, and contribution documentation, and notes that team-oriented management is handled by the related coder/coder project.

OpenCode is an open source AI coding agent aimed at developers who want AI assistance for code-related work. It provides an interactive agent experience and includes multiple built-in agents for different tasks, such as a default build agent for full-access development work and a plan agent for read-only analysis and code exploration. The project can be installed via a shell install script, common package managers, or built from source, and it is also available as a beta desktop application for macOS, Windows, and Linux. The README points users to documentation for configuration and to a docs page describing the agent system, suggesting it is meant both as a command-line tool and as a desktop companion for coding workflows.

goose is a native open source AI agent designed to run on a user’s machine. It is presented as a general-purpose assistant for code and non-code tasks such as research, writing, automation, and data analysis. The project targets people who want a local desktop application, a terminal-friendly CLI, or an embeddable API, all backed by a Rust implementation for portability and performance. The README emphasizes broad ecosystem support rather than a single provider. goose can connect to more than 15 model providers and can use either API keys or existing consumer subscriptions for services like Claude, ChatGPT, and Gemini through ACP. It also supports integration with 70+ extensions via the Model Context Protocol. The project is part of the Agentic AI Foundation at the Linux Foundation, and the README points users to installation, quickstart, tutorials, troubleshooting, and custom distribution documentation.

Cline is an open source coding agent designed to work inside an IDE or from the terminal. It targets developers who want an assistant that can read a codebase, propose changes, run commands, and help fix errors while keeping the user in control through approval-based workflows. The project is organized into multiple surfaces: a CLI, a Kanban-style web task board for parallel agent work, editor integrations for VS Code and JetBrains, and an SDK for building custom agents and tools. It supports headless automation for CI/CD, scheduled agents, plugins, MCP servers, and integrations with messaging platforms, making it suitable for both interactive development and repeatable automation across projects.

Continue is an open-source coding agent aimed at developers who want AI-assisted tooling in their editor or terminal. The README presents it as available in three forms: a CLI, a VS Code extension, and a JetBrains plugin, with a recommendation to use the CLI over the JetBrains plugin. The project documentation focuses on how to configure, understand, and customize the agent. The README also notes that this repository is a final, read-only release of the Continue 2.0.0 codebase and that anonymous telemetry was removed and authentication was pulled out in that release.

omp is a coding agent that integrates deeply with an IDE-like workflow and is designed to help developers edit, search, debug, and coordinate code changes from a terminal-style interface. It is positioned as a powerful agent surface with many built-in tools, broad model support, and a focus on getting changes right on the first attempt. The project targets users who want an agent that can do more than basic chat: it can execute code in persistent Python and Bun environments, interact with language servers, drive debuggers, manage subagents, and inject stream rules when the model drifts off task. It also supports collaboration features, shell completions, and multiple installation paths across macOS, Linux, Windows, Bun, Nix, Homebrew, and one-line installer scripts.

Pi Agent Harness is a monorepo for the Pi project, centered on a self-extensible coding agent. It includes an interactive CLI, a runtime for tool calling and state management, a terminal UI library, and a unified LLM API layer that can work with multiple providers such as OpenAI, Anthropic, and Google. The project is aimed at developers building or using coding agents, especially in local or self-hosted workflows. The README emphasizes running Pi with the permissions of the launching user and suggests containerization or sandboxing for stronger isolation. It also documents development commands, test scripts, and a workflow for publishing OSS agent sessions, indicating that the project is meant to be actively developed and used in open-source coding-agent work.

bolt.diy is an open source AI coding assistant and the community-driven version of Bolt.new. It is designed for people who want to create and iterate on full-stack Node.js applications with help from large language models, while retaining the freedom to choose among many providers such as OpenAI, Anthropic, Ollama, Gemini, and others. The README emphasizes that the project is extensible and can work with additional models supported by the Vercel AI SDK. The project can be run locally, deployed with Docker, or used as a desktop application via Electron. Its workflow centers on prompt-driven coding inside the browser or desktop app, with features such as an integrated terminal, file locking, diff viewing, code rollback, project snapshots, and ZIP export. It also includes integrations for data visualization, Git operations, Supabase, MCP, and Expo app creation, making it suitable for developers experimenting with AI-assisted application building.

Onlook is an open-source, visual-first code editor aimed at helping people craft websites, prototypes, and designs with AI. It focuses on Next.js and TailwindCSS projects, letting users start from scratch, from text or image prompts, or from prebuilt templates, then edit the resulting app visually in the browser. The project is designed for builders who want a tighter connection between design and code. Its editor supports real-time preview, direct DOM editing, code-side navigation, checkpoint-based saving, branching for experimentation, and deployment-oriented features like shareable links and custom domains. The README also indicates a web-based product with local development support, backed by a stack that includes Supabase, tRPC, Bun, Docker, and multiple AI/model providers.

Eclipse Che is a Kubernetes-native cloud development environment platform aimed at enterprise teams and developers who want reproducible, portable workspaces. It places development dependencies, runtimes, a web IDE, and project code into Kubernetes pods so environments can be distributed and run wherever Kubernetes is available. The project is organized around workspace-based development. Users can start from a Git repository URL, customize workspace components, use devfiles to make environments portable, and extend the platform with custom editors and VS Code extensions. Administrators can deploy Che on their own Kubernetes clusters and configure the system through Kubernetes custom resources, making it suitable for centrally managed team environments.

Paseo is a self-hosted platform for coordinating AI coding agents across several interfaces, including desktop, mobile, web, and a command-line client. It is aimed at developers who want to run agents on their own machines and keep their existing tools, configs, and credentials in place while switching between surfaces as needed. The project centers on a local daemon that manages agent execution and exposes connectivity for clients to attach to it. The README describes support for multiple agent providers, voice interaction, device pairing, remote access over TCP or VPN, and a skill system that helps agents hand work off to one another. It also includes a Docker deployment path and a CLI workflow for headless or remote use.

OpenHands is an AI-driven development platform centered around agentic software engineering workflows. It offers multiple entry points: a composable Python SDK for building agents in code, a CLI for interactive use, and a local GUI for running agents on a laptop. The project is aimed at developers and teams that want to delegate coding tasks to AI systems while integrating with familiar tools and model providers. The README also describes a hosted cloud deployment and an enterprise self-hosted option. The cloud version adds collaboration-oriented capabilities such as multi-user support, RBAC, conversation sharing, and integrations with Slack, Jira, and Linear. For organizations, the enterprise edition can be deployed in a VPC via Kubernetes and is source-available under a separate license, while the core project remains MIT-licensed.

CloudCLI is an open source user interface for AI coding agents such as Claude Code, Cursor CLI, and Codex. It is aimed at developers who want a browser-based way to work with agent sessions, files, Git, and terminal access from desktop or mobile devices. The project can be run locally through npm or in an experimental Docker sandbox, and it also has a managed cloud offering. Its UI focuses on session management, interactive chat, file and Git exploration, browser-based workflows, and extensibility through plugins. The README also describes a companion desktop app that connects to local or remote CloudCLI environments.

T3 Code is a control surface for agent-based coding tools that runs on a user’s machine and can be operated from mobile, web, or desktop clients. It is designed for people who already use agents such as Claude Code, Codex, Cursor, Grok Build, and OpenCode, and it acts as a remote-ready interface for controlling those local tools. The project emphasizes being open and forkable, with a focus on performance and remote access. The README points readers to in-repo documentation for installation, permissions, keyboard shortcuts, project customization, remote access, synchronization, and source control integrations, and it also describes how to run the backend locally via `npx` or install the desktop app through common distribution channels.

Aider is a terminal-based AI pair programming tool designed to help developers start new projects or extend existing codebases with assistance from large language models. It is aimed at software developers who want to work directly inside their repositories while using an AI assistant to propose and apply code changes. The project emphasizes workflows that fit into normal development practices: it can map a codebase, work with many programming languages, integrate with Git, and run linting and tests after making edits. It also supports use in an IDE or editor, can incorporate images and web pages as context, and offers voice-driven interaction. The README highlights compatibility with both hosted and local models and provides a getting-started flow based on installing via Python and running the CLI with an API key.

Prime Agent is an open-source coding and research agent intended for general-purpose and long-running workflows. It is designed for users who want a persistent, programmatic environment where model-driven actions can continue across terminal sessions, rather than ending with a single chat interaction. The project centers on two main ideas: a recursive language model abstraction that treats tools and subagents as callable functions, and a continual harness that stores prompts, memories, skills, and subagent specifications as durable session state. It supports background daemon sessions, reattachment, autonomous execution within limits, and a built-in Python REPL for file operations, shell commands, and agent coordination. The README positions it for research evaluations and sustained development work.

Twinny is a Visual Studio Code extension focused on AI-assisted programming. It is aimed at developers who want inline code completion and a chat-based interface for discussing code, generating tests, getting refactoring suggestions, and producing commit messages. The extension supports multiple AI providers, including local OpenAI/Ollama-compatible APIs as well as several hosted model providers. The README emphasizes both online and offline operation, customizable endpoints, workspace embeddings for context-aware assistance, and a decentralized Symmetry network integration for P2P AI inference. It is MIT licensed and described as free and open-source.

zerostack is a Rust-based coding agent aimed at developers who want an efficient, terminal-first assistant for software work. It combines an interactive TUI with support for multiple model providers, tool use, session continuity, and configurable permissions, positioning it as a lightweight alternative to larger JS-based coding agents. The project is designed for local use from the command line and can be installed via a script, Cargo, Homebrew, or Nix. It supports optional features such as MCP integration, persistent Markdown-based memory, lifecycle hooks, advisor workflows, multimodal input, and ACP server connectivity for editor integrations. The README emphasizes performance, sandboxing, and configurability through prompts and configuration files.

Dirac is an open-source AI coding agent designed to make code changes with high precision while minimizing token usage and API cost. It targets developers who want an assistant for refactoring, file editing, and terminal-driven tasks without relying on MCP, and it emphasizes reproducible benchmark performance on real-world repositories. The project combines stable line-hash targeting, AST-aware editing, and multi-file batching to handle larger changes efficiently. It is available as both a VS Code extension and a CLI tool, supports authentication via API keys or `dirac auth`, and can run in an approval-based workflow that lets the user stay in control while the agent reads, writes, and executes commands.

Machtiani is a terminal-based code chat service aimed at developers working in large Git repositories. It indexes project history and file context so prompts can be answered with precise references to the relevant files and functions, reducing the need for broad token-heavy searching. The project is designed to run locally and can use any model provider that follows the OpenAI API format, including local runtimes such as Ollama or MLX-LM. A typical workflow involves installing the Go CLI, starting the service with Docker Compose, syncing a repository’s commits, and then chatting with the project from the command line. It is positioned as a cost-conscious companion to other CLI tools such as Codex, especially when a compact plan or code change needs to be generated from large codebases.

Type less, code more: Cody is an AI code assistant that uses advanced search and codebase context to help you write and fix code.

Crush is a terminal-based coding assistant from Charmbracelet aimed at developers who want LLM help directly in their shell workflow. It is presented as a “coding bestie” that can work with a user’s tools, code, and project context while remaining usable across major terminal environments and operating systems. The project centers on connecting to many model providers, including OpenAI, Anthropic, Groq, OpenRouter, and others, with support for custom OpenAI- or Anthropic-compatible APIs. It keeps multiple sessions and contexts per project, can switch models during a session, and can enrich prompts through LSPs and MCPs. The README also highlights flexible configuration, including local project-level settings and global config/data locations, making it suitable for developers who want an extensible, terminal-native AI coding workflow.

Claude Code is an agentic coding assistant designed for developers who want help directly in their terminal and related workflows. It is positioned as a tool that can understand a project’s codebase, assist with routine programming tasks, explain complex code, and help manage git operations through natural language commands. The project appears aimed at individual developers and teams working in local repositories, with usage extending to the terminal, IDEs, and even GitHub via @claude mentions. The README also highlights a plugin system for extending behavior with custom commands and agents, along with documentation and troubleshooting resources for setup and usage. It is backed by Anthropic and includes information about data collection and privacy safeguards.

Roo Code is a VS Code extension positioned as an AI-powered development assistant. It is aimed at developers who want help translating natural-language requests into code changes, planning work, answering questions about a codebase, and handling routine engineering tasks without leaving the editor. The project organizes its behavior into several modes, including Code, Architect, Ask, Debug, and Custom modes, so users can switch between hands-on coding, system design, troubleshooting, and team-specific workflows. It also mentions support for MCP servers, and the README links to official documentation for installation, configuration, and usage guidance. The README notes that the extension was shut down on May 15th, and points users toward alternatives.

Void is a deprecated open-source desktop application built as a fork of Visual Studio Code. It is aimed at developers who want an AI-enhanced coding environment and a reference implementation for forking or extending VS Code itself. The project emphasizes local and flexible AI workflows: it lets users use AI agents on a codebase, inspect and checkpoint changes, and connect to different models or hosts locally. The README also highlights custom services for streaming edits, background file editing, and OS file synchronization, positioning Void as both a usable editor and a technical reference for building advanced editor integrations.