Aider is an open-source AI pair programming tool that runs in the terminal and helps developers understand, edit, test, and refactor code in a local git repository.

Aider itself is open source and free to run. You may still pay for the LLM provider or model you connect to, unless you use a free or locally hosted model.

What models does Aider support?

Aider supports models from many providers, including OpenAI, Anthropic, Google, DeepSeek, and other compatible providers, plus local model setups.

Does Aider work in the terminal?

Yes. Aider is designed around a terminal workflow and works from the command line inside your project repository.

Is Aider good for beginners?

Aider can help beginners, but it is best suited to people who already understand basic programming, terminals, git, and project setup.

Aider

Open-source AI pair programmer for terminal coding, editing, and git workflows.

Visit Website

What is Aider?

TL;DR

Aider is an open-source terminal-based AI pair programmer that works with local git repositories to edit code, implement features, fix bugs, write tests, refactor projects, and use many LLM providers.

Open-source terminal workflow for AI-assisted software development.
Works across local repositories with codebase-aware context and git.
Supports many hosted and local LLM providers.

Explore with AI

About

Aider is an open-source AI pair programming tool that runs in the terminal and works directly with local git repositories. It helps developers understand an existing codebase, edit files, implement features, fix bugs, write tests, refactor code, and keep changes inside familiar git workflows. Aider is designed for developers who prefer a lightweight terminal-first workflow instead of moving into a separate AI-native IDE. It can map repositories for broader context and supports different chat modes for coding, architecture, questions, and assistance. Developers can use it for targeted edits as well as larger multi-file changes. The tool is model-flexible and can connect to many hosted and local LLM providers. That means users can choose models based on quality, speed, privacy, or cost, while paying model providers directly for usage. This flexibility is useful for developers who do not want their coding workflow locked to one AI vendor. For founders and builders, Aider is most useful when you want AI assistance without giving up a familiar terminal and git workflow. It is a strong fit for implementation, debugging, refactoring, testing, and codebase exploration, especially for technical users who value open-source tooling and model choice.

Use Cases

Repository Feature Development

Implement features across an existing codebase while keeping changes inside your normal terminal and git workflow.

Bug Fixing & Debugging

Investigate errors, trace relevant code, propose fixes, and apply targeted changes across project files.

Automated Test Writing

Create and update tests for existing code, then iterate on failures as you improve the implementation.

Codebase Refactoring

Refactor related files, clean up duplicated logic, and make larger code changes while maintaining project context.

Key Features

Terminal-based AI pair programming for local repositories
Repository mapping gives models broader project context
Code, architect, ask, and help modes for different tasks
Works with many hosted and local LLM providers

Featured Tools

Wispr Flow

Voice-First Dictation

We may earn commissions from links to support our work. Learn more.