AI pair programmer that lives in your terminal
Aider is an open-source command-line tool that lets you pair program with an LLM directly inside your local git repository. You chat in plain English about what you want changed, and Aider edits the relevant files, applies the diffs, and commits the result with a sensible message. It supports most major models, including GPT-4o, Claude, and various open-weight models through providers like Ollama, and it builds a map of your codebase using ctags so it can work intelligently across large multi-file projects rather than just the file you have open.
It's aimed at developers who want an AI coding assistant without leaving the terminal or switching to a cloud-based IDE. Typical use is adding a feature, fixing a bug, or refactoring a chunk of code by describing the change and letting Aider propose and apply the diff, run tests, and handle the commit, all while keeping the full history in git so every AI-made change is inspectable and revertible. It also supports voice input and can pull in image or web content as context for a request.
What sets it apart from many AI coding tools is how tightly it integrates with git and how model-agnostic it is: you're not locked into one vendor's assistant, and every change lands as a normal commit you can diff, squash, or roll back like any other. That git-native workflow, combined with being free and open source, has made it a favorite among developers who want full control over their codebase and history rather than delegating that to a hosted IDE.
Lorenzo Greco
Not perfect, but it gets the job done.
Nomvula Nkosi
Refactoring a messy module took minutes instead of an afternoon. It catches bugs I would've missed for hours.
Seoyoon Park
Expected more honestly.
Christian Cruz
Picked this up for a side project and kept using it. The suggestions feel like they were written by someone who read the whole repo. Gets confused on larger files and starts suggesting irrelevant stuff.
David Cheruiyot
Picked this up for a side project and kept using it. Nothing groundbreaking, just reliable. It's become part of my daily workflow at this point.
Manuela Fernandes
The suggestions feel like they were written by someone who read the whole repo.
Baris Kurt
Been using this for about a month. Does the basics well, the extra features are hit or miss. Refactoring a messy module took minutes instead of an afternoon.
Phuc Truong
Tried it after a coworker recommended it. Not perfect, but it gets the job done. It's become part of my daily workflow at this point.
Yen Dinh
Slows down noticeably on bigger repos. Gets confused on larger files and starts suggesting irrelevant stuff.
Charlie Evans
Decent, nothing special.
Samantha Smith
Switched over from a competitor a couple months ago. Cut my debugging time down a lot this month.
Ahmed Javed
Exactly what I needed.
It's easier when you're signed in — Altern helps you get more out of AI.
By continuing you agree to our Terms and Privacy Policy.