↓ Skip to main content
9 – 17 UHR +49 8031 3508270 LUITPOLDSTR. 9, 83022 ROSENHEIM
DE / EN
Coding agent Aider-AI (Paul Gauthier) Terminal (CLI)

Aider

Aider is the open-source AI pair programmer for the terminal: Git auto-commits, a repo map and almost any model via LiteLLM, including custom endpoints.

Licence Apache-2.0
First release June 2023 (first version on PyPI)
Interfaces Terminal (CLI), Browser (--gui), Editor or IDE via watch mode
Platforms macOS, Linux, Windows
Instruction file CONVENTIONS.md (via --read)

The seven building blocks in Aider

How Aider implements the building blocks of an AI harness

  1. 1

    Agent loop

    Aider works as a chat in the terminal: you add files with /add, the model returns changes in an edit format such as diff or whole file, and Aider applies and commits them. Headless, --message or --message-file runs exactly one task.

  2. 2

    Tools and execution environment

    Instead of freely callable tools or MCP, Aider offers fixed functions: file editing, Git commands, suggested shell commands, linters and tests (--lint-cmd, --test-cmd, --auto-test), web pages and images as context, and voice input.

  3. 3

    Context management

    Per the documentation, a repo map summarizes the most important classes, functions and signatures of the whole repository and goes to the model with every request. You keep conventions in a file such as CONVENTIONS.md and load it read-only with /read.

  4. 4

    State and memory

    Aider logs the chat history in .aider.chat.history.md and reloads it with --restore-chat-history; Git holds the actual working state with one commit per change.

  5. 5

    Permissions

    Aider edits only files in the chat and, per its source code, asks before creating new files, changing other files or running suggested shell commands. --yes-always confirms prompts automatically, but per the source code not shell commands.

  6. 6

    Verification and feedback

    After every change, Aider lints by default (--auto-lint) and can run tests (--auto-test) to fix the errors it finds right away. You review changes with /diff and roll back Aider's last commit with /undo.

  7. 7

    Orchestration

    There are no subagents. Architect mode splits the work across two models: an architect model designs the solution, and an editor model turns it into concrete file edits.

Models and model access

Which models the harness uses and how they are obtained

Models and model access
Models No fixed default model: you choose it with --model or in .aider.conf.yml; the documentation points to its own LLM leaderboards for the choice
Access routes Anthropic, OpenAI, Google Gemini, DeepSeek, Azure OpenAI, Amazon Bedrock, Vertex AI, OpenRouter, Ollama, further providers via LiteLLM, custom OpenAI-compatible endpoints
Free choice of model Yes
Custom endpoint / LLM gateway Yes
Per the documentation, Aider connects to any OpenAI-compatible endpoint via OPENAI_API_BASE (or --openai-api-base) and OPENAI_API_KEY; the model name then takes the prefix openai/. Aider connects to all other providers through the LiteLLM package.
Features
Headless / CI IDE integration Checkpoints Resume sessions
Running it with herdr
Supported by herdr No
State detection Not documented
Integration None
The herdr documentation (version 0.9.5, retrieved on 22 November 2026) lists Aider neither among the integrations supported by herdr nor among those supported by the agent itself.
More about herdr →

Enterprise use

Data processing, controls and pricing model according to the vendor

Data processing

Aider runs locally and sends the files in the chat, the repo map and the history to the chosen model. For commit messages, per the documentation, diffs and chat history also go to the weak model, which may be a different provider if configured that way.

Telemetry

Per the documentation, analytics are opt-in and anonymous: without a setting, Aider asks only a random subset of users. It records models, tokens, edit formats and errors, no code and no chats. --analytics-disable turns it off permanently.

Git behavior

Aider commits every change automatically and, per the documentation, skips pre-commit hooks (--no-verify). If you want to enforce hooks such as secret scanners, set --git-commit-verify; --no-auto-commits turns auto-commits off.

Identity and access

Aider has no user or role management. Credentials live in environment variables, in .env or in .aider.conf.yml; central policies can only be enforced through distributed configuration files or an upstream gateway.

Pricing and maintenance

Aider is free under Apache-2.0, and costs arise only with the model provider. The latest version 0.86.2 was released on 12 February 2026 per PyPI, the last commit landed on 22 May 2026 per GitHub.

Strengths
  • Tight Git integration: every change as a traceable commit, rollback via /undo
  • The repo map provides compact context even in large repositories
  • Almost any model via LiteLLM, plus custom OpenAI-compatible endpoints via OPENAI_API_BASE
  • Architect mode combines a planning model with an editing model
  • Automatic linting and testing with self-correction
  • Works with any editor via comments ending in AI! or AI? in watch mode
Limitations
  • No commits since 22 May 2026, latest version 0.86.2 from 12 February 2026
  • Newer model profiles such as GPT-5.5 or Claude Opus 4.7 exist only on the main branch, not in a released version
  • No MCP, no subagents and no hooks for custom tools
  • No sandbox; suggested shell commands run with the user's permissions after confirmation
  • Pre-commit hooks are skipped by default
  • Not currently supported by herdr

Use cases

Typical tasks for this harness

Pair programming in the terminal with files you deliberately add to the chat
Traceable AI changes as individual Git commits for code reviews
Scripted changes across many files with --message in loops or CI
Comparing models on coding tasks thanks to broad LiteLLM support
Working with local models via Ollama or your own OpenAI-compatible server

Aider is an open-source AI pair programmer for the terminal that lets language models work directly in your Git repository. As an AI harness, Aider edits files, commits every change automatically and gives the model a repo map as context; through LiteLLM it connects to almost any model. As of 22 November 2026.

What is Aider?

Aider started in 2023 as a project by Paul Gauthier, who per the GitHub API has made by far the most contributions with more than 12,600 commits. The first version appeared on PyPI in June 2023; the Aider-AI/aider repository is licensed under Apache-2.0 and has 50,221 stars per the GitHub API (as of 22 November 2026). Per the installation guide, you install Aider with python -m pip install aider-install and aider-install, which brings its own Python 3.12 if needed; alternatively via uv or pipx. Then you run aider in your project directory.

Aider runs in the terminal, optionally in the browser (--gui), and in any editor via watch mode: with --watch-files, Aider reacts to code comments ending in AI! (change) or AI? (question).

Is Aider still being developed?

Barely. The latest published version is 0.86.2 from 12 February 2026 per PyPI; the latest release object on GitHub is even v0.86.0 from 9 August 2025. The last commit on the main branch dates from 22 May 2026 per the GitHub API, and the repository has been quiet since. Between February and May 2026, the additions were mainly model profiles, for example for GPT-5.5 and Claude Opus 4.7, which per the release history exist only on the main branch. Factor this maintenance status into any new rollout in your organization.

Aider as an AI harness

Aider deliberately implements the building blocks of a harness differently from autonomous agents: instead of freely callable tools and MCP, it offers fixed functions. You add the files to change with /add, the model returns changes in an edit format, and Aider applies them. Per the documentation on chat modes, Aider starts in code mode; ask answers questions without changes, architect lets one model design the solution and a second editor model write the file edits, and help explains Aider itself.

Git integration is the core: per the Git documentation, Aider commits every change with a message in the Conventional Commits style, first commits any existing uncommitted changes separately, and rolls back its own last commit with /undo. The repo map summarizes the repository’s most important classes, functions and signatures and goes to the model with every request. After changes, Aider lints by default and can run tests to fix errors itself. Headless, --message runs a single task and suits scripts.

Models and model access

Aider connects to almost every provider through the LiteLLM package: Anthropic for Claude, OpenAI for GPT, Google for Gemini, DeepSeek, Azure, Bedrock, Vertex AI, OpenRouter and local models via Ollama. The Aider leaderboards show which models edit reliably. For unknown models without a profile, Aider shows warnings but still works.

Custom endpoints are supported per the documentation on OpenAI-compatible APIs: you set OPENAI_API_BASE and OPENAI_API_KEY and start Aider with --model openai/<model-name>. This lets you run Aider in proxy mode through the AI Gateway, which checks permissions and budgets and assigns every token to a cost center.

Enterprise use: data protection and governance

Aider runs locally and sends the files in the chat, the repo map and the history to the chosen model; for commit messages, diffs and chat history go to the weak model. Per the telemetry documentation, analytics are opt-in and anonymous, contain no code and can be turned off permanently with --analytics-disable; without a setting, Aider asks only a random subset of users.

Two defaults are worth knowing. First, per the documentation, Aider skips pre-commit hooks for auto-commits; if you want to enforce secret scanners or formatters, set --git-commit-verify. Second, there is no sandbox: suggested shell commands run with the user’s permissions after confirmation, and per the source code even --yes-always does not confirm them automatically. There is no user or role management; you enforce central policies through distributed configuration files or an upstream gateway. Source code and prompts belong in your record of processing activities.

Running Aider with herdr and Orca

Aider is not currently supported by herdr: the herdr agent list in version 0.9.5 does not name Aider. Orca lists Aider in its own agent overview with automatic setup, without status reporting or hooks. Because Aider commits every change, it fits Orca’s approach of one Git worktree per task well.

How does Aider differ from Claude Code and Codex?

Aider is a pair programmer, while Claude Code and Codex work as autonomous agents with tools, MCP servers and subagents. With Aider, you decide which files the model sees and changes, and every change lands as a Git commit. Aider is model-agnostic, whereas Claude Code and Codex come from the model vendors Anthropic and OpenAI. If you are looking for a model-agnostic, actively maintained agent with MCP, OpenCode is one option.

Where it fits: developer harness and enterprise harness

Aider is a harness for developers who work in the terminal and with Git. For business users without a command line, CompanyGPT is the matching harness, with MCP tools, enterprise knowledge and roles from Entra ID or Keycloak. The AI Gateway provides the shared cost and governance layer; per the product page, its companion measures agent time for Claude Code, Codex, OpenCode, Pi, GitHub Copilot CLI and Gemini CLI, so you capture Aider through proxy mode. Our overview of AI harnesses covers further coding agents and orchestrators.

From developer harness to enterprise harness

Coding harnesses are tools for developers. Business users, cost control and governance need two more layers.

CompanyGPT – the harness for business users

CompanyGPT gives any language model tools via MCP servers, integrations and n8n, company knowledge via companyRAG, roles from Entra ID or Keycloak, and audit logs – operated in your own cloud tenant (Azure, AWS or Google Cloud) or sovereignly on STACKIT.

Discover CompanyGPT
AI Gateway – cost and governance for every agent

The AI Gateway bundles the model access of your tools and coding agents behind an OpenAI-compatible API: sign-in via Entra ID, Keycloak or Cognito, budgets and hard limits per team and user, cost attribution per project and repository via the Companion, and logging of every request.

View the AI Gateway

Frequently Asked Questions

What is Aider?

Aider is an open-source AI pair programmer for the terminal, developed by Paul Gauthier and published under Apache-2.0. You add files to the chat, describe the change, and Aider edits the code directly in your Git repository and commits every change with a suitable message. A repo map gives the model an overview of the entire project.

Is Aider open source and free?

Yes. Aider is licensed under Apache-2.0 and is free; you install it via pip, uv or pipx. Costs arise only with the model provider, for example by token usage. Note the maintenance status: the latest version 0.86.2 was released on 12 February 2026 per PyPI, and the last commit in the repository landed on 22 May 2026 per GitHub.

Which models does Aider support?

Aider connects to almost every provider through the LiteLLM package, including Anthropic, OpenAI, Google Gemini, DeepSeek, Azure, Bedrock, Vertex AI and OpenRouter, plus local models via Ollama. The Aider leaderboards show which models edit well. Profiles for very recent models currently sit partly on the main branch only, but models also work without a profile, with warnings.

Does Aider work with a custom endpoint or the innFactory AI Gateway?

Yes. Per the documentation, you set OPENAI_API_BASE to the endpoint, store the key in OPENAI_API_KEY and start Aider with the model name prefixed with openai/. This runs Aider in proxy mode through the AI Gateway, which checks budgets and assigns tokens to a cost center. The gateway companion's agent time measurement does not currently cover Aider.

Can Aider be used in a GDPR-compliant way?

That depends on the model provider. Aider itself runs locally, and analytics are opt-in and contain no code. The files in the chat, the repo map and, for commit messages, also diffs do go to the configured models. These providers must be covered contractually and listed in your record of processing activities; EU endpoints, your own gateway or local models keep the data under control.

How does Aider differ from Claude Code and Codex?

Aider is a pair programmer: you decide which files the model edits, and every change lands as a Git commit. Claude Code and Codex work as autonomous agents with tools, MCP and subagents and come from the model vendors Anthropic and OpenAI. Aider is model-agnostic but does without MCP, subagents and a sandbox; since May 2026 there have been no commits in the repository.

Introduce coding agents safely in your company

We support selection, model access in the EU, permission design and cost control.