Product / UX / Orchestration & agents

Aider

Terminal-based AI pair programming tool that edits code directly in your local git repo. Differentiates through deep git integration (automatic commits with sensible messages), support for 20+ LLM providers, and a sophisticated diff-based editing approach that minimizes token usage. 45K GitHub stars; widely regarded as the benchmark for CLI-based coding agents and the tool that pioneered the 'agentic coding in terminal' pattern.

Aider v0.86.2 (PyPI aider-chat, Feb 12 2026); GitHub repo Aider-AI/aider active, 45.8k stars. Terminal AI pair-programming agent that plans edits across a repo and applies git commits. Confirmed live June 2026.

Openness

5 high confidence

5.0

license: Apache-2.0(OSI)
source: public(GitHub)
core-gated: ungated

Verified Apache-2.0 OSI, full source public, no managed/enterprise tier gating the core. Repo owner Aider-AI/aider, 45.8k stars confirmed via primary GitHub source.

https://github.com/Aider-AI/aider recorded 2026-06-04
Apache-2.0 license, public source, 45.8k stars (verified)
https://pypi.org/project/aider-chat/ recorded 2026-06-04
aider-chat v0.86.2 published Feb 12 2026

Adoption

3 medium confidence

3.0

GitHub 'used by' ~605 dependent projects and broad terminal-CLI install base; 45.8k stars corroborate but are not the basis. No published download/user count in the millions, so placed at 3 on dependent/usage-volume signal.

https://github.com/Aider-AI/aider recorded 2026-06-04
~605 dependent repos ('used by')

Capability

3 medium confidence

3.0

Mid-tier on SWE-bench Verified; well below OpenHands (68.4%) and Cline (59.8%). NOTE: headline figure sourced from aggregator leaderboard (awesomeagents.ai), not a primary SWE-bench publication; kept at medium confidence.

https://awesomeagents.ai/leaderboards/swe-bench-coding-agent-leaderboard/ recorded 2026-06-04
Aider (architect mode) SWE-bench Verified 31.4% (aggregator)

Unchanged since 2026-07-30 (last edited, not re-checked)