Run an Army of Coding Agents
The open-source workspace for Claude Code, Codex, and any coding agent.
Start work from anywhere
Mobile, Cloud, or bring your own Cloud
Trusted by builders from
Switch agents. Keep your workflow
Use Claude Code, Codex, OpenCode, or any CLI agent. Your workspaces, branches, and review stay the same.
Scale to 100+ agents
Run parallel coding agents on features, fixes, and refactors. Each one works on its own branch, locally or in the cloud.
Start agents on events or a schedule
Agents triage issues, draft changelogs, and bump dependencies on their own.
Drive it from the terminal
Create workspaces and start agents from the CLI, the SDK, or over MCP.
What builders say about Paradev
Private by default.
You're in control.
Local workspaces run on your machine. You decide what connects to them.
Open Source
Paradev is MIT licensed, so you're never locked in. Read or change the code on GitHub.
View on GitHubLocal First
In local workspaces, your code and agent sessions stay on your machine. Cloud workspaces run in a sandbox.
Your Agents, Your Accounts
Use the agent plans and API keys you already have. Model calls go straight to your provider.
Frequently
asked questions
How is Paradev different from just running Claude Code in a terminal?
Claude Code, Codex, and OpenCode are the agents; Paradev is where you run many of them at once. Each task gets its own Git worktree and branch, so ten agents can work on ten branches simultaneously while you monitor, review, and merge from one place.
See how Paradev compares to other tools →I already use an IDE like Cursor, is this for me?
Paradev is designed to work with your existing tool, we natively support deep-linking to IDEs like Cursor so you can open your workspaces and files in your IDE.
Which AI coding agents are supported?
Paradev works with any CLI-based coding agent, including Claude Code, OpenCode, OpenAI Codex, and more. Choose a different agent for every task without changing your workspace or review flow.
How does the parallel agent system work?
Every agent runs in its own Git worktree and branch, keeping working files separate. Changes can still conflict when merged. You watch, review, and merge them all from one window.
Read the guide to parallel coding agents →Is Paradev free to use?
Paradev has a free tier. The source code is on GitHub under the MIT License, so you can inspect, modify, and self-host it.
Can I use my own API keys?
Absolutely. Paradev doesn't proxy any API calls. You use your own API keys directly with whatever AI providers you choose. This means you have full control over costs and usage.
Is Paradev open source?
Yes. Paradev is open source under the MIT License. The code is public on GitHub, and anyone can file issues, open pull requests, or fork it. Paradev is unrelated to Apache Superset, the business-intelligence tool.
Read the MIT License →What platforms does Paradev run on?
The desktop app runs on macOS, with an experimental Linux AppImage; Windows is not yet available. Beyond the desktop app there's a CLI, a TypeScript SDK, and an MCP server, so you can drive Paradev from scripts, terminals, and other agents. An app for iPhone and iPad on the App Store lets you check on agents and review their diffs away from your desk.
Is Paradev just a wrapper around Claude Code?
No. Your agents stay independent; Paradev is the orchestration layer around them. Each task gets its own Git worktree and branch, persistent sessions, diff review, and automations that run on events or a schedule, without tying the workflow to one provider.