ModelCompare
Compare

AI coding agent usage

What was the most used AI coding agent on ?

Claude Code from Anthropic, with a usage score of 46 out of 100 across 3 of 3 sources. OpenClaw was next at 39.

Usage score over time, week by week

Usage ranking

AI coding agents by usage on March 1, 2026

This ranks how much each agent is used and followed, not how good it is: a popular agent can be a worse fit for your work than a small one. Click an agent to see where its numbers come from. Each source's column shows the agent's scaled score (the leader is 100) over the raw number. 3 sources had data on this day. The score averages an agent's scaled values over the sources that can measure it (at least three).

  1. 1
    Claude Code
    Anthropic · closed source
    Usage score 46
    • npm 100 · 9.3M
    • GitHub 30 · 72k
    • VS Code 9 · 5.4M
    • npm: 1009,342,431 downloads of @anthropic-ai/claude-code in the 7 days to Mar 1, 2026. That's the most of any agent in this source, so it scores 100. Source
    • GitHub: 3071,847 stars on github.com/anthropics/claude-code on Mar 1, 2026. The leader here is OpenClaw with 241k, so Claude Code scores 72k ÷ 241k × 100 = 30. Source
    • VS Code: 95,362,769 installs of the anthropic.claude-code extension on Mar 1, 2026. The leader here is GitHub Copilot with 59M, so Claude Code scores 5.4M ÷ 59M × 100 = 9. Source

    Usage score: (100 + 30 + 9) ÷ 3 = 46. Claude Code website

  2. 2
    OpenClaw
    OpenClaw · open source
    Usage score 39
    • npm 16 · 1.5M
    • GitHub 100 · 241k
    • npm: 161,479,089 downloads of openclaw in the 7 days to Mar 1, 2026. The leader here is Claude Code with 9.3M, so OpenClaw scores 1.5M ÷ 9.3M × 100 = 16. Source
    • GitHub: 100240,811 stars on github.com/openclaw/openclaw on Mar 1, 2026. That's the most of any agent in this source, so it scores 100. Source

    Not counted: VS Code, because it has no VS Code extension.

    Usage score: (16 + 100) ÷ 3 = 39, divided by 3 rather than 2 because every agent's average counts at least 3 sources. OpenClaw website

  3. 3
    GitHub Copilot
    GitHub · closed source
    Usage score 34
    • npm 2 · 188k
    • VS Code 100 · 59M
    • npm: 2187,858 downloads of @github/copilot in the 7 days to Mar 1, 2026. The leader here is Claude Code with 9.3M, so GitHub Copilot scores 188k ÷ 9.3M × 100 = 2. Source
    • VS Code: 10059,257,804 installs of the GitHub.copilot-chat extension on Feb 2, 2026. That's the most of any agent in this source, so it scores 100. Source

    Not counted: GitHub, because it has no public GitHub repo.

    Usage score: (2 + 100) ÷ 3 = 34, divided by 3 rather than 2 because every agent's average counts at least 3 sources. GitHub Copilot website

  4. 4
    Codex CLI
    OpenAI · open source
    Usage score 19
    • npm 23 · 2.1M
    • GitHub 26 · 62k
    • VS Code 8 · 5M
    • npm: 232,140,172 downloads of @openai/codex in the 7 days to Mar 1, 2026. The leader here is Claude Code with 9.3M, so Codex CLI scores 2.1M ÷ 9.3M × 100 = 23. Source
    • GitHub: 2662,454 stars on github.com/openai/codex on Mar 1, 2026. The leader here is OpenClaw with 241k, so Codex CLI scores 62k ÷ 241k × 100 = 26. Source
    • VS Code: 85,036,019 installs of the openai.chatgpt extension on Mar 1, 2026. The leader here is GitHub Copilot with 59M, so Codex CLI scores 5M ÷ 59M × 100 = 8. Source

    Usage score: (23 + 26 + 8) ÷ 3 = 19. Codex CLI website

  5. 5
    Gemini CLI
    Google · open source
    Usage score 16
    • npm 8 · 759k
    • GitHub 40 · 96k
    • VS Code 1 · 584k
    • npm: 8758,678 downloads of @google/gemini-cli in the 7 days to Mar 1, 2026. The leader here is Claude Code with 9.3M, so Gemini CLI scores 759k ÷ 9.3M × 100 = 8. Source
    • GitHub: 4096,161 stars on github.com/google-gemini/gemini-cli on Mar 2, 2026. The leader here is OpenClaw with 241k, so Gemini CLI scores 96k ÷ 241k × 100 = 40. Source
    • VS Code: 1583,958 installs of the Google.gemini-cli-vscode-ide-companion extension on Feb 17, 2026. The leader here is GitHub Copilot with 59M, so Gemini CLI scores 584k ÷ 59M × 100 = 1. Source

    Usage score: (8 + 40 + 1) ÷ 3 = 16. Gemini CLI website

  6. 6
    OpenCode
    Anomaly · open source
    Usage score 15
    • npm 5 · 473k
    • GitHub 39 · 94k
    • npm: 5473,386 downloads of opencode-ai in the 7 days to Mar 1, 2026. The leader here is Claude Code with 9.3M, so OpenCode scores 473k ÷ 9.3M × 100 = 5. Source
    • GitHub: 3994,012 stars on github.com/anomalyco/opencode on Feb 1, 2026. The leader here is OpenClaw with 241k, so OpenCode scores 94k ÷ 241k × 100 = 39. Source

    Not counted: VS Code, because there's no snapshot of its extension page from around that day.

    Usage score: (5 + 39) ÷ 3 = 15, divided by 3 rather than 2 because every agent's average counts at least 3 sources. OpenCode website

  7. 7
    Cline
    Cline · open source
    Usage score 10
    • npm 1 · 139k
    • GitHub 24 · 58k
    • VS Code 5 · 3.1M
    • npm: 1139,208 downloads of cline in the 7 days to Mar 1, 2026. The leader here is Claude Code with 9.3M, so Cline scores 139k ÷ 9.3M × 100 = 1. Source
    • GitHub: 2457,537 stars on github.com/cline/cline on Feb 3, 2026. The leader here is OpenClaw with 241k, so Cline scores 58k ÷ 241k × 100 = 24. Source
    • VS Code: 53,058,125 installs of the saoudrizwan.claude-dev extension on Feb 9, 2026. The leader here is GitHub Copilot with 59M, so Cline scores 3.1M ÷ 59M × 100 = 5. Source

    Usage score: (1 + 24 + 5) ÷ 3 = 10. Cline website

  8. 8
    Roo Code
    Roo Code · open source
    Usage score 4
    • GitHub 9 · 22k
    • VS Code 2 · 1.2M
    • GitHub: 922,086 stars on github.com/RooCodeInc/Roo-Code on Feb 3, 2026. The leader here is OpenClaw with 241k, so Roo Code scores 22k ÷ 241k × 100 = 9. Source
    • VS Code: 21,238,301 installs of the RooVeterinaryInc.roo-cline extension on Feb 9, 2026. The leader here is GitHub Copilot with 59M, so Roo Code scores 1.2M ÷ 59M × 100 = 2. Source

    Not counted: npm, because it has no command-line package on npm.

    Usage score: (9 + 2) ÷ 3 = 4, divided by 3 rather than 2 because every agent's average counts at least 3 sources. Roo Code website

  9. 9
    Qwen Code
    Alibaba · open source
    Usage score 3
    • npm 1 · 58k
    • GitHub 8 · 20k
    • npm: 158,417 downloads of @qwen-code/qwen-code in the 7 days to Mar 1, 2026. The leader here is Claude Code with 9.3M, so Qwen Code scores 58k ÷ 9.3M × 100 = 1. Source
    • GitHub: 819,738 stars on github.com/QwenLM/qwen-code on Mar 2, 2026. The leader here is OpenClaw with 241k, so Qwen Code scores 20k ÷ 241k × 100 = 8. Source

    Not counted: VS Code, because it has no VS Code extension.

    Usage score: (1 + 8) ÷ 3 = 3, divided by 3 rather than 2 because every agent's average counts at least 3 sources. Qwen Code website

  10. 10
    Crush
    Charm · open source
    Usage score 3
    • npm 0 · 3.4k
    • GitHub 9 · 21k
    • npm: 03,420 downloads of @charmland/crush in the 7 days to Mar 1, 2026. The leader here is Claude Code with 9.3M, so Crush scores 3.4k ÷ 9.3M × 100 = 0. Source
    • GitHub: 920,724 stars on github.com/charmbracelet/crush on Mar 2, 2026. The leader here is OpenClaw with 241k, so Crush scores 21k ÷ 241k × 100 = 9. Source

    Not counted: VS Code, because it has no VS Code extension.

    Usage score: (0 + 9) ÷ 3 = 3, divided by 3 rather than 2 because every agent's average counts at least 3 sources. Crush website

  11. 11
    Kilo Code
    Kilo · open source
    Usage score 3
    • npm 0 · 31k
    • GitHub 7 · 16k
    • VS Code 1 · 710k
    • npm: 030,720 downloads of @kilocode/cli in the 7 days to Mar 1, 2026. The leader here is Claude Code with 9.3M, so Kilo Code scores 31k ÷ 9.3M × 100 = 0. Source
    • GitHub: 716,057 stars on github.com/Kilo-Org/kilocode on Mar 1, 2026. The leader here is OpenClaw with 241k, so Kilo Code scores 16k ÷ 241k × 100 = 7. Source
    • VS Code: 1710,118 installs of the kilocode.Kilo-Code extension on Feb 8, 2026. The leader here is GitHub Copilot with 59M, so Kilo Code scores 710k ÷ 59M × 100 = 1. Source

    Usage score: (0 + 7 + 1) ÷ 3 = 3. Kilo Code website

  12. 12
    pi
    Mario Zechner · open source
    Usage score 1
    • npm 0 · 7.6k
    • GitHub 2 · 4.5k
    • npm: 07,588 downloads of @mariozechner/pi-coding-agent in the 7 days to Mar 1, 2026. The leader here is Claude Code with 9.3M, so pi scores 7.6k ÷ 9.3M × 100 = 0. Source
    • GitHub: 24,478 stars on github.com/badlogic/pi-mono on Feb 1, 2026. The leader here is OpenClaw with 241k, so pi scores 4.5k ÷ 241k × 100 = 2. Source

    Not counted: VS Code, because it has no VS Code extension.

    Usage score: (0 + 2) ÷ 3 = 1, divided by 3 rather than 2 because every agent's average counts at least 3 sources. pi website

  13. 13
    Auggie
    Augment Code · closed source
    Usage score 1
    • npm 0 · 37k
    • VS Code 1 · 700k
    • npm: 036,773 downloads of @augmentcode/auggie in the 7 days to Mar 1, 2026. The leader here is Claude Code with 9.3M, so Auggie scores 37k ÷ 9.3M × 100 = 0. Source
    • VS Code: 1699,841 installs of the augment.vscode-augment extension on Feb 9, 2026. The leader here is GitHub Copilot with 59M, so Auggie scores 700k ÷ 59M × 100 = 1. Source

    Not counted: GitHub, because it has no public GitHub repo.

    Usage score: (0 + 1) ÷ 3 = 1, divided by 3 rather than 2 because every agent's average counts at least 3 sources. Auggie website

  14. 14
    Amp
    Sourcegraph · closed source
    Usage score 0
    • npm 0 · 36k
    • VS Code 0 · 99k
    • npm: 035,613 downloads of @sourcegraph/amp in the 7 days to Mar 1, 2026. The leader here is Claude Code with 9.3M, so Amp scores 36k ÷ 9.3M × 100 = 0. Source
    • VS Code: 098,538 installs of the sourcegraph.amp extension on Feb 19, 2026. The leader here is GitHub Copilot with 59M, so Amp scores 99k ÷ 59M × 100 = 0. Source

    Not counted: GitHub, because it has no public GitHub repo.

    Usage score: (0 + 0) ÷ 3 = 0, divided by 3 rather than 2 because every agent's average counts at least 3 sources. Amp website

  15. 15
    Letta Code
    Letta · open source
    Usage score 0
    • npm 0 · 4.7k
    • GitHub 0 · 1k
    • npm: 04,672 downloads of @letta-ai/letta-code in the 7 days to Mar 1, 2026. The leader here is Claude Code with 9.3M, so Letta Code scores 4.7k ÷ 9.3M × 100 = 0. Source
    • GitHub: 01,024 stars on github.com/letta-ai/letta-code on Feb 12, 2026. The leader here is OpenClaw with 241k, so Letta Code scores 1k ÷ 241k × 100 = 0. Source

    Not counted: VS Code, because it has no VS Code extension.

    Usage score: (0 + 0) ÷ 3 = 0, divided by 3 rather than 2 because every agent's average counts at least 3 sources. Letta Code website

Not ranked that day, only one source has them: OpenHands (GitHub, 67k), Aider (GitHub, 41k), Hermes Agent (GitHub, 137), Droid (npm, 829).

Method

How the usage score is made

No single public number measures how much each AI coding agent is used, so we blend the ones that exist. In each source the leader scores 100 and the rest are scaled to it. An agent's score is the average over the sources that can measure it. Sources that can't see an agent at all (Cursor has no npm package) are left out of its average, but the average always counts at least three sources, so an agent seen by only one or two can't outrank one that leads everywhere. An agent needs two sources to be ranked.

  • npmCLI downloads, 7 days · data from Mar 1, 2026

    Downloads of each agent's command-line package in the 7 days to the date, with one-off bursts (mirrors, bots) capped. Downloads count installs and CI runs, not people, and agents with their own installer are undercounted. Source

  • GitHubStars · data from Mar 2, 2026

    Stars on the agent's GitHub repo, from monthly snapshots. Stars measure interest, and build up over time. Source

  • VS CodeExtension installs, all time · data from Mar 1, 2026

    All-time installs of the agent's VS Code extension, from monthly snapshots. Cursor and Windsurf install from Open VSX instead, which isn't counted. Source

Browse

The most used AI coding agent, month by month

2026
2025

Pick the model behind your agent

Most of these agents let you choose the model. Send one prompt to up to six models and see which answer you'd actually use, with what each one cost.

Compare models

Data through October 21, 2026 for 20 agents. npm downloads from the npm registry; OpenRouter app rankings from OpenRouter (CC BY 4.0); GitHub stars and VS Code installs from monthly snapshots in the Internet Archive and the live APIs; survey figures as published by Stack Overflow and JetBrains.