AI Agent Index
ByHeather MacAvelia·Independently reviewed·Published Mar 18, 2026·Updated Sep 24, 2026
Independently verified against live vendor data on Sep 24, 2026.

Open-source AI pair programming tool that runs in your terminal and commits each edit to git. Free, and you bring your own model API key.

How we scored it

Autonomy

4/5

Integrations

3/5

Pricing clarity

5/5

Evidence

3/5

Setup

5/5

The facts

Aider is an open-source AI pair programming tool that runs in the terminal and edits code inside a git repository. It is released under the Apache 2.0 license, and its GitHub repository has 49.1k stars. The project's website reports 6.8 million installs and 15 billion tokens processed per week, which are the project's own figures. The tool is free and has no paid tier. Users bring their own API key for a model provider, or run a local model, and pay that provider directly for usage. The documentation covers OpenAI, Anthropic, Gemini, DeepSeek, xAI, Groq, Cohere, Azure, Amazon Bedrock, Vertex AI, OpenRouter, Ollama, LM Studio and other OpenAI-compatible APIs. The project publishes no estimate of typical monthly API cost, which depends on the model chosen and how much it is used. Each time Aider edits a file it commits the change to git with a descriptive commit message, so changes can be reviewed with ordinary git tools or reversed with the /undo command. Auto-commits can be switched off. It builds a map of the whole repository to give the model context in larger codebases, runs a built-in linter after each edit and offers to fix the errors it finds, and can run a project's tests after each change when that option is turned on. Other features include voice input, adding images and web pages to the chat, and a watch mode that picks up instructions written as code comments in any IDE or editor. The project's leaderboards compare how different models perform when editing code with Aider. Cursor ($16/mo billed annually) and GitHub Copilot ($10/mo) work inside an editor with inline suggestions, while Aider works from the terminal. OpenCode and Claude Code are other terminal coding agents: OpenCode is also open source and works with many model providers, while Claude Code ($17/mo billed annually) comes with Anthropic's paid plans. The project offers no commercial support, SLA or compliance certification for the tool itself, and help runs through GitHub issues and the project's Discord. How code and prompts are handled depends on the model provider a team connects. Analytics are opt-in and anonymous, and the project states they never collect code, chat messages or keys. The most recent commit on the GitHub repository is dated May 22, 2026, and the latest tagged GitHub release is v0.86.0 from August 9, 2025. The documentation does not cover MCP.

Pricing

free

Segment

b2c

Setup

moderate

Verified

Sep 24, 2026

Transparency

Public

Contract

—

Data training

Not Trained

Human in loop

Optional

Capabilities

git-nativeterminalopen-sourceversion-control

Pros & Limitations

Editorial assessment

Pros

  • ✓Free and open source under the Apache 2.0 license, so the code can be audited, forked and run without a vendor account.
  • ✓Works with many model providers, or a local model, through your own API key, so you pay only the provider's usage charges and can switch models at any time.
  • ✓Commits every edit to git with a descriptive message, so AI changes can be reviewed with normal git tools or undone with the /undo command.
  • ✓Builds a map of the whole repository and runs a linter after each edit, and a watch mode takes instructions written as code comments from any IDE or editor.

Limitations

  • ⚠Terminal-first: using it comfortably takes familiarity with the command line, git and managing model API keys.
  • ⚠No commercial support, SLA or compliance certification for the tool itself, and data handling depends on the model provider you connect.
  • ⚠Model usage is billed by your provider, so spending and rate limits have to be watched and managed with that provider.
  • ⚠The latest tagged GitHub release is v0.86.0 from August 2025, and the most recent commit to the repository is from May 2026.

Technical Details

Deployment
cli
Model architectureModel-agnostic
Avg setup timeNo published setup duration. Install with the aider-install package, add an API key for a model provider, then run aider inside a git repository.
Autonomous rateApplies edits and commits them to git without a per-edit approval by default (auto-commits can be switched off), and the --yes-always option answers yes to every remaining confirmation.
Integrations
Git (any repository)Any IDE or editor (watch mode)

Similar agents

Rating

3.6/ 5
★★★★★

Editorial score

How we score this →

Recognition

Transparent PricingListed 2026
Embed these on your site →

Built for

Open SourceDevToolsStartups

Leave a review

Never displayed publicly.

Agent Stacks

See workflow stacks that feature Aider.

Compare

Related Content

Manage this listing

Badges, review collection, and listing updates for verified vendors.

Vendor dashboard →

Free · Every Two Weeks

AI Agent Price & Rating Tracker

Price changes, new agent launches, acquisitions, and rating updates across the AI agent market. Verified against live vendor data, not vendor marketing.

No spam. Unsubscribe anytime. We never share your email.