Aider
AI pair programming in your terminal
What is Aider?
Aider is a command-line AI coding assistant launched in April 2023 by Paul Gauthier. It runs from the terminal and integrates seamlessly with Git, enabling developers to interact with large language models for writing, refactoring, and testing code. Aider maps the codebase, sends relevant context to an LLM, applies edits as diffs, and commits changes automatically with sensible messages. It supports both cloud and local models, over 100 programming languages, voice-to-code input, and lint/test automation.
What you can do with it
Bootstrapping new projects
Scaffold project structure, boilerplate and core features via conversational prompts.
Cross‑file refactoring
Apply coordinated changes across multiple files in a single refactoring session.
Test‑driven development
Generate tests and adjust code to pass testing workflows automatically.
Legacy code modernization
Update deprecated patterns or migrate frameworks with atomic Git changes.
Documentation and commenting
Automatically generate docstrings, comments, or module documentation for clarity.
Key features
- Terminal‑first AI pair programming
- Whole‑repository context mapping (repo map)
- Support for both cloud and local LLMs (BYO‑model)
- Multi‑file edits with Git auto‑commits
- Multiple chat modes (code, architect, ask, help)
- Voice‑to‑code, image and webpage context support
- Automated linting and testing after edits
Screenshots
Inputs / Outputs
Strengths & Limitations
Strengths
Open-source and free to use
Licensed under Apache 2.0; no subscription required.
Git-aware editing workflow
Automatically applies AI edits as diffs and commits with generated commit messages, preserving version history.
Flexible model support (BYOK)
Works with many cloud LLMs and local models; users pay only for tokens consumed.
Rich input capabilities
Supports voice input, images, web pages, AI comments in code, providing varied context to the model.
Broad language & IDE integration
Handles over 100 languages; works from terminal and integrates with IDEs through comment-based triggers.
Automated linting and testing
Runs linters and tests after every edit, and can automatically fix issues.
Limitations
Terminal-first interface
No standalone GUI; may have steeper learning curve for non-command-line users.
Dependent on external LLM APIs
Users must manage API keys and model costs; heavy usage can become expensive.
Setup complexity
Requires initial configuration and familiarity with CLI tools and environment variables.
Large repo context challenges
While repo mapping helps, large codebases may still strain context size or performance.
Pricing & Plans
Model: Freemium
Free / Open‑source
Full-featured CLI tool under Apache 2.0 with no subscription or usage limits
BYO‑LLM API usage
You pay only the underlying LLM provider’s usage fees, no markup from Aider
Local model usage
Run with local LLMs for zero per‑request cost besides compute
Aider is free and open-source (Apache 2.0); users supply their own LLM API keys and pay per token to the chosen LLM provider (typical monthly cost $10–$100 depending on usage and model). Local models like Ollama can be used at no cost beyond hardware.
Who it's for
Ideal for
Experienced developers comfortable with CLI and Git who want full control over AI-assisted code editing and LLM costs.
Not ideal for
Users preferring GUI tools, minimal setup, or integrated IDE plugins with fixed pricing.
What users say
- Highly productive for coding workflows
- Open-source community praised
- Appreciated transparency of costs
- Terminal-based attracts developers
- Flexibility with model choice valued
- Setup and usage feels advanced for less technical users
Prompts & Results
›Refactor this function to use async/await patterns. AI!
Aider sends the diff suggestion to LLM, which rewrites the function accordingly and commits the changes with a message like 'refactor: convert function to async/await'
›Write a unit test for the calculate_sum function. AI?
Aider adds a new test file with assertions covering edge cases and commits with message 'test: add unit tests for calculate_sum'
›Voice-to-code: fix the off-by-one error in the loop.
Aider interprets spoken instruction, applies the fix in code, and commits with message 'fix: correct off-by-one in loop'
›Add comments to all public methods for documentation. AI!
Aider inserts docstrings or comments for each public method and commits with message 'docs: add comments to public methods'
FAQ
Is Aider free to use?+
Yes—Aider itself is open-source and free; you only pay for the LLM API tokens you use, or no cost if using local models.
Which LLMs does Aider support?+
It supports a wide range including Claude Sonnet, GPT-4o, DeepSeek, Gemini, and local models via OpenRouter or Ollama.
How does Aider apply changes to my code?+
Aider generates diff-style edits and applies them via Git commits, with autogenerated commit messages for traceability.
Can I use Aider with my IDE?+
Yes—you can trigger Aider using special comments (AI!, AI?) in code files that Aider watches when run with --watch-files.
Ratings & Reviews
No reviews yet — be the first to rate this tool.