Open-source terminal coding agent that pairs with git, edits multiple files at once, and works with any leading LLM.
Open-source terminal coding agent that pairs with git, edits multiple files at once, and works with any leading LLM.
Aider is one of the original terminal-native AI pair programmers and remains a favorite of developers who want a lightweight, git-first workflow. You launch aider inside a repo, tell it what to change, and it reads the relevant files, proposes an edit as a diff, applies it, and automatically commits the change with a descriptive message. Because every step is a git commit, undoing anything is a simple git reset. Aider supports diff, whole-file, and unified-diff edit formats and picks the right one per model to maximize edit success rate.
Aider is model-agnostic. You bring your own key for Anthropic Claude, OpenAI GPT-5, Google Gemini, xAI Grok, DeepSeek, Groq, Fireworks, Together, OpenRouter, or any OpenAI-compatible endpoint including self-hosted Ollama or vLLM. It maintains an ongoing published leaderboard of models against a code-editing benchmark, which has become one of the most-cited comparisons in the community. Notable features include voice-to-code, repo-map indexing for large codebases, /architect mode for reasoning-first planning, and an experimental MCP client so Aider can call external tools. The project is Apache-2.0 licensed, has no commercial upsell, and is entirely community-maintained. It is the tool most experienced engineers reach for when they want a terminal agent that does one thing (edit code with git) extremely well.
Was this helpful?
The strongest free AI coding tool for terminal-native developers who want full model choice. Competes with paid alternatives on real-world benchmarks but requires comfort with command-line workflows and careful API cost management.
Free
Ready to get started with Aider?
View Pricing Options →We believe in transparent reviews. Here's what Aider doesn't handle well:
Weekly insights on the latest AI tools, features, and trends delivered to your inbox.
Homepage highlights support for Claude 3.7 Sonnet, DeepSeek R1 & Chat V3, OpenAI o1, and o3-mini — all 2025-era frontier models. Usage stats reported: 44K GitHub stars, 6.8M installs, 15 billion tokens processed per week, and an 88% 'singularity' score (percentage of Aider's own code written by Aider). No explicit 2026 roadmap is published on the homepage.
Coding Agents
AI-first code editor with autonomous coding capabilities. Understands your codebase and writes code collaboratively with you.
AI Coding Assistant
Open-source AI coding agent for VS Code that gives you full model choice, transparent cost tracking, and native MCP support.
AI Coding Assistant
Open-source AI code assistant for VS Code and JetBrains — bring your own models, curate a shared block library, ship as an internal platform.
AI Coding IDE
Agentic coding IDE (now part of Cognition/Devin) that pairs the SWE-1 model family with Cascade multi-step agents for large-scale refactors.
AI Agent Builders
Codeium has rebranded to Devin Desktop under Cognition — an IDE plus coding-agent command center with Free, Pro $20, Max $200, and Team plans.
No reviews yet. Be the first to share your experience!
Get started with Aider and see if it's the right fit for your needs.
Get Started →Take our 60-second quiz to get personalized tool recommendations
Find Your Perfect AI Stack →Explore 20 ready-to-deploy AI agent templates for sales, support, dev, research, and operations.
Browse Agent Templates →Compare the top AI coding agents in 2026 — Claude Code, Cursor, Copilot, Codex, Windsurf, Aider, and more. Real pricing, honest strengths, and a decision framework for every skill level.
Two years ago, picking an AI coding assistant meant choosing between a handful of autocomplete plugins. In 2026, the **cursor vs github copilot** decision affects how your entire team writes, reviews, and ships code. Both tools have added agentic capabilities, background task run