Loading the page.
Topic hub
index/AI agents
updated 8 November 2026AI coding agents
Programs that read a repository, edit files, run tests, and take a task through to a commit on their own. The whole landscape - on one map: tile size - strength in our ranking, color - vendor camp.
Size = our score · click - profileOpenAIAnthropicGoogleCognitionCursor
players in the field
18
CLI trio installs / week
40.6M
week over week
+10%
growing fastest
Claude Code +10%
live race of the week - analysis below
Race of the week
Who's driving the market right now
CLI agents were downloaded 40.6M times this week - 10% more than last week. Codex is confidently first (25.9M/week vs. 14.2M/week for Claude Code); Gemini CLI - a distant third (477K/week).
Claude Code14.2M/week
Codex25.9M/week
Gemini CLI477K/week
api.npmjs.org · data points through Nov 8 · updated hourly · CLI trio only: the others use different distribution channels
how the CLI trio's week is split (npm)
Claude Code35%
Codex64%
Gemini CLI1%
2.0
Our score is the tool’s overall strength: benchmarks, development activity, ecosystem. It's not the same as popularity: compare it with the race above.
Codex
97.0
Claude Code
91.0
Cursor
90.0
Devin
78.0
Windsurf
77.0
Copilot Agent
69.0
Jules
65.0
Antigravity
64.0
3.0
Catalog: 18 agents
A profile is your entry point: pricing, limits, benchmarks, track record, all with verification dates.
Codex
97.0OpenAI
OpenAI's coding agent for code and computer work: desktop app, CLI (open source), IDE, web/cloud, GitHub review, SDK and control from your phone; models - GPT-6 Sol and Luna (since Sep 22) and the top-tier GPT-6 Astra.
Claude Code
91.0Anthropic
Anthropic AI agent for repository-scale tasks: a terminal-first process with auto mode, dynamic workflows, subagents, background agents, and control through diff, tests, limits, and usage.
Cursor
90.0Cursor
Cursor - an editor for AI development: best where work happens with eyes and hands in the IDE, not in a separate cloud runner.
Devin
78.0Cognition
Devin - a cloud contractor for tasks: give it an issue, get a PR attempt, and review it as work from an external contributor.
Windsurf
77.0Cognition
Windsurf - an IDE agent at the intersection of the editor and a managed workflow: assess it with fresh eyes, not memories of old Codeium.
Copilot Agent
69.0GitHub
GitHub Copilot Coding Agent - an executor inside GitHub: turns an issue into a branch/PR if the team knows how to write issues.
Jules
65.0Google Jules - an async runner for GitHub tasks: yes for small assignments, no for a full workstation.
Antigravity
64.0Google Antigravity - an experimental agent-first IDE: the direction is interesting, but maturity must be proven on a branch.
Replit Agent
51.0Replit
Web-based app builder within Replit: suitable for fast prototypes and beginners, but does not replace a serious local coding workflow.
Lovable
43.0Lovable
Lovable - a fast builder for demos and MVP screens: it sells speed to product, not engineering maturity.
Aider
40.0Aider
Aider - an honest open-source CLI: less polish, more control; you choose and pay for the model yourself.
Bolt
36.0StackBlitz
Bolt - a browser-first builder: quickly create a web prototype, then move it into a proper engineering process.
Junie
32.0JetBrains
JetBrains Junie - an AI agent for JetBrains teams: if the IDE is already chosen, test it; if not, this is not the first decision point.
Augment
28.0Augment
Augment Code - an enterprise agent for large context: needed where the repository is already bigger than one developer's head.
OpenCode
20.0SST
OpenCode - an open agent harness: for those who want provider control and do not want a closed default.
Cline
13.0Cline
Cline - an open IDE agent with approvals and checkpoints: control is good, but responsibility for configuration remains with the user.
ChatGPT
—OpenAI
OpenAI's universal chat assistant: GPT-5.6 in chat (Luna on Free/Go, Sol on paid plans), GPT-6 Sol and Luna in the Work agent mode and in Codex since Sep 22, the top-end GPT-6 Astra - from Plus; voice, images and ready-made artifacts.
Gemini
—Google's proprietary AI assistant powered by Gemini models (3.8 Flash and 3.1 Pro): three models without a subscription, up to 1M-token context on Pro/Ultra, Deep Research, voice, and Omni Flash video; built into Search, Workspace, Android, and Chrome.
4.0
Pairs: agent + model
An agent's score is always measured as an "agent + model" pair. In the AA Coding Agent Index as of Nov 9, 2026, Claude Code's best result is with Opus 5.5, and Codex's with GPT-6 Astra. Compare pairs, not brands.
← Claude Code + Opus 5.5Codex + GPT-6 Astra →
AA Coding Agent IndexNov 9, 2026
61.7
54.7
Rope tilt = who is stronger on this benchmark; bright side - winner
5.0
What's new in the field
Events from the profile track records of two flagships: releases and fixes push up, incidents and limit policy push down.
2026-09-22
Claude Code: Opus 5.5 in Claude Code (v2.1.280): 1M context window, $4/$20 per 1M tokens
2026-09-22
Codex: GPT-6 Sol and GPT-6 Luna in Codex: half the API price of GPT-5.6
2026-09-04
Codex: GPT-6 Astra in Codex: OpenAI's top model, on the Plus, Pro and Business plans
2026-09-01
Claude Code: Fable 5.1 in Claude Code (v2.1.257): same price of $10/$50, cache reads 75% cheaper
2026-08-25
Codex: Plus gets back the 5-hour Codex limit window, Pro still lives without it
6.0
How to choose for yourself
Pick not the "best overall" but the one that fits your task.
Complex work in your own repo
Claude Code →
Affordable pipeline for repetitive tasks
Codex →
Open client with your own key
Cline →
Which AI coding agent should you choose in 2026?
For complex work in your own repo, take Claude Code: in the AA Coding Agent Index as of Sep 25, 2026 it holds the top two spots, with Opus 5.5 and with Fable 5.1. Codex with GPT-6 Astra scores lower, but it solves a task twice as fast as the leader and at almost half the price on API rates.
Are there free agents?
Yes. Codex on the free ChatGPT plan works in the desktop app with the GPT-6 Luna model. The open Cline and OpenCode cost nothing, you pay only for the model with your own key. Claude Code has no free plan: entry starts at $17 a month when billed annually.
Agent and model: what is the difference?
The model comes up with the solution, the agent carries it out: reads the repo, edits files, runs tests and makes a commit. One agent works with different models, so the score is measured as a pair. Claude Code with Opus 5.5 and with Fable 5.1 have different scores, price and task time. Compare pairs, not brands.
7.0
Where to go next
The hub - an entry point into the topic; depth lives in profiles, concepts, and practice.
Try it yourself
- Guides and practice
- MCP server catalog for agents
- Token maximization: get the most from subscriptions
