Aider logo

Aider

AI pair programming in your terminal with LLMs

Aider enables developers to pair program with large language models directly from the command line, working seamlessly with existing codebases and supporting 100+ programming languages.

Aider banner

Overview

Terminal-Native AI Coding Assistant

Aider brings AI pair programming directly into your terminal, letting you collaborate with leading LLMs like Claude 3.7 Sonnet, DeepSeek R1, and OpenAI o3-mini to build new projects or enhance existing codebases. Unlike browser-based coding assistants, Aider integrates deeply with your development workflow through native Git integration and codebase mapping.

Built for Real-World Development

Aider automatically maps your entire codebase to provide context-aware suggestions across large projects. It supports over 100 programming languages and commits changes with sensible messages, making it easy to review, diff, and roll back AI-generated code using standard Git workflows. The tool integrates with your existing IDE, accepts voice commands, and can process images and web pages for visual context.

Developers can leverage automatic linting and testing after each change, ensuring code quality without manual intervention. With 3.4M+ installations and processing 15B tokens weekly, Aider has proven itself in production environments. The project itself is 88% self-coded, demonstrating the tool's capability to handle complex, real-world software development tasks.

Highlights

Codebase mapping for context-aware suggestions across large projects
Native Git integration with automatic commits and sensible messages
Support for 100+ programming languages and multiple LLM providers
Voice-to-code, image context, and automatic linting/testing workflows

Pros

  • Works directly in terminal with existing development workflows
  • Supports both cloud LLMs and local models for flexibility
  • Automatic Git commits enable easy review and rollback of changes
  • IDE integration and voice commands enhance accessibility

Considerations

  • Command-line interface may have learning curve for GUI-focused developers
  • Optimal performance requires premium LLM API access
  • Codebase mapping effectiveness depends on project structure
  • Terminal-based workflow may not suit all development preferences

Managed products teams compare with

When teams consider Aider, these hosted platforms usually appear on the same shortlist.

Amazon Q Developer logo

Amazon Q Developer

Generative AI coding assistant for building, operating, and transforming software

Claude Code logo

Claude Code

AI pair‑programmer for code generation, refactors, and explanations

CodeGPT logo

CodeGPT

AI code assistant for generating, explaining, and refactoring code

Looking for a hosted option? These are the services engineering teams benchmark against before choosing open source.

Fit guide

Great for

  • Developers comfortable with terminal-based workflows and Git
  • Teams working on large codebases requiring context-aware assistance
  • Engineers seeking AI coding help without leaving their IDE
  • Projects requiring multi-language support and flexible LLM options

Not ideal when

  • Developers preferring exclusively GUI-based coding assistants
  • Teams unable to integrate API-based LLM services
  • Projects with strict restrictions on external AI tool usage
  • Users seeking no-configuration, browser-only coding solutions

How teams use it

Legacy Codebase Modernization

Map existing code and request refactoring, test coverage, or feature additions with full project context

Rapid Prototyping with Voice Commands

Speak feature requests and let Aider implement changes while you review and iterate hands-free

Cross-Language Development

Work seamlessly across Python, JavaScript, Rust, and 100+ languages in polyglot projects

Test-Driven Development Acceleration

Automatically lint and test AI-generated code, with Aider fixing detected issues in real-time

Tech snapshot

Python80%
CSS4%
Shell4%
Tree-sitter Query4%
JavaScript4%
HTML3%

Tags

llamasonnetcommand-linegpt-3geminianthropicgpt-4oclichatgptclaude-3gpt-4openaigpt-35-turbo

Frequently asked questions

Which LLMs does Aider support?

Aider works best with Claude 3.7 Sonnet, DeepSeek R1 & Chat V3, OpenAI o1, o3-mini, and GPT-4o, but can connect to almost any LLM including local models.

How does Aider handle Git integration?

Aider automatically commits changes with sensible commit messages, allowing you to use familiar Git tools to diff, manage, and undo AI-generated changes.

Can I use Aider with my existing IDE?

Yes, Aider integrates with your favorite IDE or editor. You can request changes by adding comments to your code and Aider will implement them.

What programming languages does Aider support?

Aider supports 100+ programming languages including Python, JavaScript, Rust, Ruby, Go, C++, PHP, HTML, CSS, and dozens more.

How does codebase mapping work?

Aider creates a map of your entire codebase to provide context-aware assistance, helping it work effectively in larger projects by understanding code relationships and structure.

Project at a glance

Active
Stars
39,909
Watchers
39,909
Forks
3,825
LicenseApache-2.0
Repo age2 years old
Last commit2 days ago
Primary languagePython

Last synced yesterday