Skip to main content

Aider Logo

AI Pair Programming in Your Terminal

Aider lets you pair program with LLMs to start a new project or build on your existing codebase.

aider screencast

GitHub Stars PyPI Downloads Tokens per week OpenRouter Ranking Singularity

This is ForgePair, a fork of aider-AI/aider under different governance (see SPEC.md for why). The stats/badges above are inherited from upstream aider and reflect its install base and history, not ForgePair's -- ForgePair is a young fork with its own CHANGES.md. See STATUS.md for what's done, what's known-incomplete, and where to look for more detail before relying on this fork for anything beyond experimentation.

What's different from aider

ForgePair keeps aider's terminal-first, git-native, model-agnostic core (see SPEC.md §3 for what was deliberately kept as-is) and changes two things: how the project is governed, and a handful of specific, long-requested capabilities upstream never shipped.

Governance -- so shipped work doesn't sit unreleased for months:

  • CI is a real hard merge gate on main (tests across 5 Python versions x 2 OSes, lint, Docker build), not advisory -- confirmed by a direct push being rejected.
  • A live merge queue (GitHub's native merge queue, org-owned repo) processes approved, green PRs automatically, without the maintainer personally clicking merge each time.
  • Continuous release: every merge to main auto-computes the next version and publishes to PyPI via Trusted Publishing (OIDC) -- no stored long-lived token. This directly targets the exact failure mode found in upstream aider at fork time (commits existed, but the last PyPI release was 3+ months older than the last commit).
  • Tiered contributor categories (low-risk: docs/model-metadata/ dependency bumps vs. requires deeper review) documented in CONTRIBUTING.md.

Features aider users have been requesting since 2024, with nowhere to go (see SPEC.md §2 for the sourcing):

  • Native MCP (Model Context Protocol) client support (--edit-format agent) -- tools, resources, and prompts, plus authenticated remote HTTP MCP servers (bearer token / API key).
  • Approval-gated apply mode (--confirm-edits) -- diffs are gated before applying, not just shown after the fact.
  • Honest failure handling on provider errors -- no more silent 0-token responses on rate-limits/API failures.
  • Issue/request triage that treats recurrence as a severity signal instead of dedup-closing reports into silence, and gives feature requests the same triage attention bug reports get.

See STATUS.md for what's actually verified-done today vs. still in progress, and CHANGES.md for the dated changelog.

Features

Cloud and local LLMs

Aider works best with Claude 3.7 Sonnet, DeepSeek R1 & Chat V3, OpenAI o1, o3-mini & GPT-4o, but can connect to almost any LLM, including local models.


Maps your codebase

Aider makes a map of your entire codebase, which helps it work well in larger projects.


100+ code languages

Aider works with most popular programming languages: python, javascript, rust, ruby, go, cpp, php, html, css, and dozens more.


Git integration

Aider automatically commits changes with sensible commit messages. Use familiar git tools to easily diff, manage and undo AI changes.


Use in your IDE

Use aider from within your favorite IDE or editor. Ask for changes by adding comments to your code and aider will get to work.


Images & web pages

Add images and web pages to the chat to provide visual context, screenshots, reference docs, etc.


Voice-to-code

Speak with aider about your code! Request new features, test cases or bug fixes using your voice and let aider implement the changes.


Linting & testing

Automatically lint and test your code every time aider makes changes. Aider can fix problems detected by your linters and test suites.


Copy/paste to web chat

Work with any LLM via its web chat interface. Aider streamlines copy/pasting code context and edits back and forth with a browser.

Getting Started

python -m pip install aider-install
aider-install

# Change directory into your codebase
cd /to/your/project

# DeepSeek
aider --model deepseek --api-key deepseek=<key>

# Claude 3.7 Sonnet
aider --model sonnet --api-key anthropic=<key>

# o3-mini
aider --model o3-mini --api-key openai=<key>

See the installation instructions and usage documentation for more details.

More Information

Documentation

Community & Resources

Kind Words From Users

Metadata

Release files for forgepair 1.0.3

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for forgepair 1.0.3
File Size Uploaded
forgepair-1.0.3.tar.gz 1.4 MB Details

Built distribution (wheel)

Table of built distributions (wheels) for forgepair 1.0.3
File Interpreter ABI Platform
forgepair-1.0.3-py3-none-any.whl Python 3 none any Details

Total release size: 1.8 MB

Release files / forgepair-1.0.3.tar.gz

Download URL forgepair-1.0.3.tar.gz
Size 1.4 MB
Tags Source
SHA-256 checksum
How to use checksums
5324c7c34813098f424a63b286d36351d39846023cf3679d32a0b7f643154a9f
BLAKE2b-256 checksum
How to use checksums
873cfcbeaa60e66ac4cfbcb2d2fb3fe7f59c594675c3ee7264c5c41f50524f79
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Sep 16, 2026.

Transparency log

Release files / forgepair-1.0.3-py3-none-any.whl

Download URL forgepair-1.0.3-py3-none-any.whl
Size 400.7 kB
Tags Python 3
SHA-256 checksum
How to use checksums
a57ef3a84aaf95629c11652e8e4b087b26e33ecb11cd0458a6f1c4b99401b4ba
BLAKE2b-256 checksum
How to use checksums
3a8b0a2d4ed35f5175c91f72b203eba9d4eb9d9c75ce1336d716c936a0b9ee35
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Sep 16, 2026.

Transparency log

Release history Release notifications | RSS feed

This release

1.0.3 This release

2 release files

1.0.2

2 release files

1.0.1

2 release files

1.0.0

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page