Aider

EDITOR PICK
aider.chat ↗

AI pair programming in your terminal, with git commits and voice input

SI Score
72/100
Pricing
Not confirmed
Free plan
Not confirmed
Sign-up required
Not confirmed

Ask SI about Aider

Answers use saved listing facts and recorded sources.
Homepage screenshot
Screenshot of the Aider homepage

Aider is a command-line tool for pair programming with LLMs. It maps your codebase, edits files in your local git repo and commits changes with generated messages. You install it with pip, point it at a model such as DeepSeek, Claude or o3-mini, and chat about the changes you want.

Checked against the site's own pages on .

Our take

Worth trying if you already live in a terminal and use git daily. The git integration is the practical draw: every AI edit becomes a commit you can diff or undo. Expect to supply your own API key, and note the docs say it works best with specific hosted models rather than any LLM equally.

Features

  • Git integration

    Aider commits its changes automatically with commit messages and you can diff, manage or undo them with normal git tools.

  • Codebase map

    It builds a map of your entire codebase, which the docs say helps on larger projects.

  • Voice input

    The /voice command records your spoken request, transcribes it and treats it as a typed chat instruction.

  • Lint and test

    Aider runs your linters and tests after changes and can fix problems they report.

Best for

  • Developers who work in a terminal
  • Teams already using git for every change
  • Projects large enough to need a codebase map
  • People who want to pick their own LLM provider

Not for

  • Developers who want a graphical editor only

FAQ

Does Aider work with local LLMs?
Yes. The feature list says it can connect to almost any LLM, including local models, though it works best with Claude 3.7 Sonnet, DeepSeek R1 and Chat V3, OpenAI o1, o3-mini and GPT-4o.
How do I install Aider?
The getting-started section shows python -m pip install aider-install followed by aider-install, then running aider from inside your project directory with a model and API key.

Sources