ModelCompare
Compare

AI coding agent usage

What was the most used AI coding agent on ?

Claude Code from Anthropic, with a usage score of 70 out of 100 across 6 of 6 sources. GitHub Copilot was next at 62.

Usage score over time, week by week

Usage ranking

AI coding agents by usage on November 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. 6 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 70
    • npm 46 · 13M
    • OpenRouter 100 · 40T
    • GitHub 39 · 153k
    • VS Code 36 · 29M
    • JetBrains 100 · 39%
    • Stack Overflow 100 · 66%
    • npm: 4612,886,034 downloads of @anthropic-ai/claude-code in the 7 days to Nov 1, 2026. The leader here is Codex CLI with 28M, so Claude Code scores 13M ÷ 28M × 100 = 46. Source
    • OpenRouter: 10040T tokens sent through OpenRouter by Claude Code in the week to Oct 23, 2026. That's the most of any agent in this source, so it scores 100. Source
    • GitHub: 39153,146 stars on github.com/anthropics/claude-code on Oct 23, 2026. The leader here is OpenClaw with 393k, so Claude Code scores 153k ÷ 393k × 100 = 39. Source
    • VS Code: 3628,525,807 installs of the anthropic.claude-code extension on Oct 23, 2026. The leader here is GitHub Copilot with 79M, so Claude Code scores 29M ÷ 79M × 100 = 36. Source
    • JetBrains Developer Ecosystem 2026: 10039% of 15,000 respondents said they use it. Question: AI coding tools used at work, May to July 2026. That's the most of any agent in this source, so it scores 100. Source
    • Stack Overflow Developer Survey 2026: 10065.5% of 12,255 respondents said they use it. Question: Coding agents and assistants used in the past year. That's the most of any agent in this source, so it scores 100. Source

    Usage score: (46 + 100 + 39 + 36 + 100 + 100) ÷ 6 = 70. Claude Code website

  2. 2
    GitHub Copilot
    GitHub · closed source
    Usage score 62
    • npm 6 · 1.8M
    • VS Code 100 · 79M
    • JetBrains 54 · 21%
    • Stack Overflow 90 · 59%
    • npm: 61,784,936 downloads of @github/copilot in the 7 days to Nov 1, 2026. The leader here is Codex CLI with 28M, so GitHub Copilot scores 1.8M ÷ 28M × 100 = 6. Source
    • VS Code: 10078,946,959 installs of the GitHub.copilot-chat extension on Oct 23, 2026. That's the most of any agent in this source, so it scores 100. Source
    • JetBrains Developer Ecosystem 2026: 5421% of 15,000 respondents said they use it. Question: AI coding tools used at work, May to July 2026. The leader here is Claude Code with 39%, so GitHub Copilot scores 21% ÷ 39% × 100 = 54. Source
    • Stack Overflow Developer Survey 2026: 9058.7% of 12,255 respondents said they use it. Question: Coding agents and assistants used in the past year. The leader here is Claude Code with 66%, so GitHub Copilot scores 59% ÷ 66% × 100 = 90. Source

    Not counted: OpenRouter, because it doesn't show up in OpenRouter's app rankings; GitHub, because it has no public GitHub repo.

    Usage score: (6 + 100 + 54 + 90) ÷ 4 = 62. GitHub Copilot website

  3. 3
    OpenClaw
    OpenClaw · open source
    Usage score 46
    • npm 9 · 2.6M
    • OpenRouter 28 · 11T
    • GitHub 100 · 393k
    • npm: 92,578,610 downloads of openclaw in the 7 days to Nov 1, 2026. The leader here is Codex CLI with 28M, so OpenClaw scores 2.6M ÷ 28M × 100 = 9. Source
    • OpenRouter: 2811T tokens sent through OpenRouter by OpenClaw in the week to Oct 23, 2026. The leader here is Claude Code with 40T, so OpenClaw scores 11T ÷ 40T × 100 = 28. Source
    • GitHub: 100392,864 stars on github.com/openclaw/openclaw on Oct 23, 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; JetBrains Developer Ecosystem 2026, because the survey didn't list it; Stack Overflow Developer Survey 2026, because the survey didn't list it.

    Usage score: (9 + 28 + 100) ÷ 3 = 46. OpenClaw website

  4. 4
    Codex CLI
    OpenAI · open source
    Usage score 45
    • npm 100 · 28M
    • OpenRouter 30 · 12T
    • GitHub 33 · 131k
    • VS Code 20 · 16M
    • JetBrains 41 · 16%
    • Stack Overflow 45 · 30%
    • npm: 10027,839,068 downloads of @openai/codex in the 7 days to Nov 1, 2026. That's the most of any agent in this source, so it scores 100. Source
    • OpenRouter: 3012T tokens sent through OpenRouter by Codex CLI in the week to Oct 23, 2026. The leader here is Claude Code with 40T, so Codex CLI scores 12T ÷ 40T × 100 = 30. Source
    • GitHub: 33130,619 stars on github.com/openai/codex on Oct 23, 2026. The leader here is OpenClaw with 393k, so Codex CLI scores 131k ÷ 393k × 100 = 33. Source
    • VS Code: 2015,867,909 installs of the openai.chatgpt extension on Oct 23, 2026. The leader here is GitHub Copilot with 79M, so Codex CLI scores 16M ÷ 79M × 100 = 20. Source
    • JetBrains Developer Ecosystem 2026: 4116% of 15,000 respondents said they use it. Question: AI coding tools used at work, May to July 2026. The leader here is Claude Code with 39%, so Codex CLI scores 16% ÷ 39% × 100 = 41. Source
    • Stack Overflow Developer Survey 2026: 4529.5% of 12,255 respondents said they use it. Question: Coding agents and assistants used in the past year. The leader here is Claude Code with 66%, so Codex CLI scores 30% ÷ 66% × 100 = 45. Source

    Usage score: (100 + 30 + 33 + 20 + 41 + 45) ÷ 6 = 45. Codex CLI website

  5. 5
    OpenCode
    Anomaly · open source
    Usage score 25
    • npm 19 · 5.2M
    • GitHub 56 · 221k
    • VS Code 2 · 1.3M
    • npm: 195,155,514 downloads of opencode-ai in the 7 days to Nov 1, 2026. The leader here is Codex CLI with 28M, so OpenCode scores 5.2M ÷ 28M × 100 = 19. Source
    • GitHub: 56220,529 stars on github.com/anomalyco/opencode on Oct 23, 2026. The leader here is OpenClaw with 393k, so OpenCode scores 221k ÷ 393k × 100 = 56. Source
    • VS Code: 21,259,159 installs of the sst-dev.opencode extension on Oct 23, 2026. The leader here is GitHub Copilot with 79M, so OpenCode scores 1.3M ÷ 79M × 100 = 2. Source

    Not counted: OpenRouter, because it doesn't show up in OpenRouter's app rankings; JetBrains Developer Ecosystem 2026, because the survey didn't list it; Stack Overflow Developer Survey 2026, because the survey didn't list it.

    Usage score: (19 + 56 + 2) ÷ 3 = 25. OpenCode website

  6. 6
    Cursor
    Anysphere · closed source
    Usage score 23
    • JetBrains 31 · 12%
    • Stack Overflow 39 · 26%
    • JetBrains Developer Ecosystem 2026: 3112% of 15,000 respondents said they use it. Question: AI coding tools used at work, May to July 2026. The leader here is Claude Code with 39%, so Cursor scores 12% ÷ 39% × 100 = 31. Source
    • Stack Overflow Developer Survey 2026: 3925.5% of 12,255 respondents said they use it. Question: Coding agents and assistants used in the past year. The leader here is Claude Code with 66%, so Cursor scores 26% ÷ 66% × 100 = 39. Source

    Not counted: npm, because it has no command-line package on npm; OpenRouter, because it doesn't show up in OpenRouter's app rankings; GitHub, because it has no public GitHub repo; VS Code, because it has no VS Code extension.

    Usage score: (31 + 39) ÷ 3 = 23, divided by 3 rather than 2 because every agent's average counts at least 3 sources. Cursor website

  7. 7
    Hermes Agent
    Nous Research · open source
    Usage score 22
    • OpenRouter 0 · 12B
    • GitHub 65 · 256k
    • OpenRouter: 012B tokens sent through OpenRouter by Hermes Agent in the week to Oct 23, 2026. The leader here is Claude Code with 40T, so Hermes Agent scores 12B ÷ 40T × 100 = 0. Source
    • GitHub: 65256,486 stars on github.com/NousResearch/hermes-agent on Oct 23, 2026. The leader here is OpenClaw with 393k, so Hermes Agent scores 256k ÷ 393k × 100 = 65. Source

    Not counted: npm, because it has no command-line package on npm; VS Code, because it has no VS Code extension; JetBrains Developer Ecosystem 2026, because the survey didn't list it; Stack Overflow Developer Survey 2026, because the survey didn't list it.

    Usage score: (0 + 65) ÷ 3 = 22, divided by 3 rather than 2 because every agent's average counts at least 3 sources. Hermes Agent website

  8. 8
    pi
    Mario Zechner · open source
    Usage score 21
    • npm 0 · 7.6k
    • OpenRouter 34 · 14T
    • GitHub 30 · 117k
    • npm: 07,588 downloads of @mariozechner/pi-coding-agent in the 7 days to Nov 1, 2026. The leader here is Codex CLI with 28M, so pi scores 7.6k ÷ 28M × 100 = 0. Source
    • OpenRouter: 3414T tokens sent through OpenRouter by pi in the week to Oct 23, 2026. The leader here is Claude Code with 40T, so pi scores 14T ÷ 40T × 100 = 34. Source
    • GitHub: 30117,416 stars on github.com/badlogic/pi-mono on Oct 23, 2026. The leader here is OpenClaw with 393k, so pi scores 117k ÷ 393k × 100 = 30. Source

    Not counted: VS Code, because it has no VS Code extension; JetBrains Developer Ecosystem 2026, because the survey didn't list it; Stack Overflow Developer Survey 2026, because the survey didn't list it.

    Usage score: (0 + 34 + 30) ÷ 3 = 21. pi website

  9. 9
    Kilo Code
    Kilo · open source
    Usage score 19
    • npm 0 · 60k
    • OpenRouter 66 · 26T
    • GitHub 7 · 28k
    • VS Code 2 · 1.7M
    • npm: 060,160 downloads of @kilocode/cli in the 7 days to Nov 1, 2026. The leader here is Codex CLI with 28M, so Kilo Code scores 60k ÷ 28M × 100 = 0. Source
    • OpenRouter: 6626T tokens sent through OpenRouter by Kilo Code in the week to Oct 23, 2026. The leader here is Claude Code with 40T, so Kilo Code scores 26T ÷ 40T × 100 = 66. Source
    • GitHub: 727,721 stars on github.com/Kilo-Org/kilocode on Oct 23, 2026. The leader here is OpenClaw with 393k, so Kilo Code scores 28k ÷ 393k × 100 = 7. Source
    • VS Code: 21,658,660 installs of the kilocode.Kilo-Code extension on Oct 23, 2026. The leader here is GitHub Copilot with 79M, so Kilo Code scores 1.7M ÷ 79M × 100 = 2. Source

    Not counted: JetBrains Developer Ecosystem 2026, because the survey didn't list it; Stack Overflow Developer Survey 2026, because the survey didn't list it.

    Usage score: (0 + 66 + 7 + 2) ÷ 4 = 19. Kilo Code website

  10. 10
    Cline
    Cline · open source
    Usage score 19
    • npm 0 · 104k
    • OpenRouter 49 · 20T
    • GitHub 18 · 71k
    • VS Code 7 · 5.7M
    • npm: 0103,990 downloads of cline in the 7 days to Nov 1, 2026. The leader here is Codex CLI with 28M, so Cline scores 104k ÷ 28M × 100 = 0. Source
    • OpenRouter: 4920T tokens sent through OpenRouter by Cline in the week to Oct 23, 2026. The leader here is Claude Code with 40T, so Cline scores 20T ÷ 40T × 100 = 49. Source
    • GitHub: 1870,777 stars on github.com/cline/cline on Oct 23, 2026. The leader here is OpenClaw with 393k, so Cline scores 71k ÷ 393k × 100 = 18. Source
    • VS Code: 75,711,166 installs of the saoudrizwan.claude-dev extension on Oct 23, 2026. The leader here is GitHub Copilot with 79M, so Cline scores 5.7M ÷ 79M × 100 = 7. Source

    Not counted: JetBrains Developer Ecosystem 2026, because the survey didn't list it; Stack Overflow Developer Survey 2026, because the survey didn't list it.

    Usage score: (0 + 49 + 18 + 7) ÷ 4 = 19. Cline website

  11. 11
    OpenHands
    All Hands AI · open source
    Usage score 14
    • OpenRouter 19 · 7.5T
    • GitHub 23 · 92k
    • OpenRouter: 197.5T tokens sent through OpenRouter by OpenHands in the week to Oct 23, 2026. The leader here is Claude Code with 40T, so OpenHands scores 7.5T ÷ 40T × 100 = 19. Source
    • GitHub: 2391,657 stars on github.com/OpenHands/OpenHands on Oct 23, 2026. The leader here is OpenClaw with 393k, so OpenHands scores 92k ÷ 393k × 100 = 23. Source

    Not counted: npm, because it has no command-line package on npm; VS Code, because it has no VS Code extension; JetBrains Developer Ecosystem 2026, because the survey didn't list it; Stack Overflow Developer Survey 2026, because the survey didn't list it.

    Usage score: (19 + 23) ÷ 3 = 14, divided by 3 rather than 2 because every agent's average counts at least 3 sources. OpenHands website

  12. 12
    Gemini CLI
    Google · open source
    Usage score 10
    • npm 2 · 445k
    • GitHub 27 · 108k
    • VS Code 1 · 1.1M
    • npm: 2445,269 downloads of @google/gemini-cli in the 7 days to Nov 1, 2026. The leader here is Codex CLI with 28M, so Gemini CLI scores 445k ÷ 28M × 100 = 2. Source
    • GitHub: 27107,559 stars on github.com/google-gemini/gemini-cli on Oct 23, 2026. The leader here is OpenClaw with 393k, so Gemini CLI scores 108k ÷ 393k × 100 = 27. Source
    • VS Code: 11,125,384 installs of the Google.gemini-cli-vscode-ide-companion extension on Oct 23, 2026. The leader here is GitHub Copilot with 79M, so Gemini CLI scores 1.1M ÷ 79M × 100 = 1. Source

    Not counted: OpenRouter, because it doesn't show up in OpenRouter's app rankings; JetBrains Developer Ecosystem 2026, because the survey didn't list it; Stack Overflow Developer Survey 2026, because the survey didn't list it.

    Usage score: (2 + 27 + 1) ÷ 3 = 10. Gemini CLI website

  13. 13
    Aider
    Aider · open source
    Usage score 4
    • OpenRouter 0 · 15B
    • GitHub 13 · 50k
    • OpenRouter: 015B tokens sent through OpenRouter by Aider in the week to Oct 23, 2026. The leader here is Claude Code with 40T, so Aider scores 15B ÷ 40T × 100 = 0. Source
    • GitHub: 1349,716 stars on github.com/Aider-AI/aider on Oct 23, 2026. The leader here is OpenClaw with 393k, so Aider scores 50k ÷ 393k × 100 = 13. Source

    Not counted: npm, because it has no command-line package on npm; VS Code, because it has no VS Code extension; JetBrains Developer Ecosystem 2026, because the survey didn't list it; Stack Overflow Developer Survey 2026, because the survey didn't list it.

    Usage score: (0 + 13) ÷ 3 = 4, divided by 3 rather than 2 because every agent's average counts at least 3 sources. Aider website

  14. 14
    Roo Code
    Roo Code · open source
    Usage score 3
    • OpenRouter 1 · 243B
    • GitHub 6 · 24k
    • VS Code 3 · 2.1M
    • OpenRouter: 1243B tokens sent through OpenRouter by Roo Code in the week to Oct 23, 2026. The leader here is Claude Code with 40T, so Roo Code scores 243B ÷ 40T × 100 = 1. Source
    • GitHub: 624,268 stars on github.com/RooCodeInc/Roo-Code on Oct 23, 2026. The leader here is OpenClaw with 393k, so Roo Code scores 24k ÷ 393k × 100 = 6. Source
    • VS Code: 32,098,503 installs of the RooVeterinaryInc.roo-cline extension on Oct 23, 2026. The leader here is GitHub Copilot with 79M, so Roo Code scores 2.1M ÷ 79M × 100 = 3. Source

    Not counted: npm, because it has no command-line package on npm; JetBrains Developer Ecosystem 2026, because the survey didn't list it; Stack Overflow Developer Survey 2026, because the survey didn't list it.

    Usage score: (1 + 6 + 3) ÷ 3 = 3. Roo Code website

  15. 15
    Qwen Code
    Alibaba · open source
    Usage score 3
    • npm 1 · 140k
    • OpenRouter 0 · 187B
    • GitHub 7 · 29k
    • npm: 1140,414 downloads of @qwen-code/qwen-code in the 7 days to Nov 1, 2026. The leader here is Codex CLI with 28M, so Qwen Code scores 140k ÷ 28M × 100 = 1. Source
    • OpenRouter: 0187B tokens sent through OpenRouter by Qwen Code in the week to Oct 23, 2026. The leader here is Claude Code with 40T, so Qwen Code scores 187B ÷ 40T × 100 = 0. Source
    • GitHub: 728,670 stars on github.com/QwenLM/qwen-code on Oct 23, 2026. The leader here is OpenClaw with 393k, so Qwen Code scores 29k ÷ 393k × 100 = 7. Source

    Not counted: VS Code, because it has no VS Code extension; JetBrains Developer Ecosystem 2026, because the survey didn't list it; Stack Overflow Developer Survey 2026, because the survey didn't list it.

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

  16. 16
    Crush
    Charm · open source
    Usage score 2
    • npm 0 · 16k
    • GitHub 7 · 29k
    • npm: 015,854 downloads of @charmland/crush in the 7 days to Nov 1, 2026. The leader here is Codex CLI with 28M, so Crush scores 16k ÷ 28M × 100 = 0. Source
    • GitHub: 728,770 stars on github.com/charmbracelet/crush on Oct 23, 2026. The leader here is OpenClaw with 393k, so Crush scores 29k ÷ 393k × 100 = 7. Source

    Not counted: OpenRouter, because it doesn't show up in OpenRouter's app rankings; VS Code, because it has no VS Code extension; JetBrains Developer Ecosystem 2026, because the survey didn't list it; Stack Overflow Developer Survey 2026, because the survey didn't list it.

    Usage score: (0 + 7) ÷ 3 = 2, divided by 3 rather than 2 because every agent's average counts at least 3 sources. Crush website

  17. 17
    Letta Code
    Letta · open source
    Usage score 1
    • npm 0 · 113k
    • OpenRouter 2 · 909B
    • GitHub 1 · 3.7k
    • npm: 0112,941 downloads of @letta-ai/letta-code in the 7 days to Nov 1, 2026. The leader here is Codex CLI with 28M, so Letta Code scores 113k ÷ 28M × 100 = 0. Source
    • OpenRouter: 2909B tokens sent through OpenRouter by Letta Code in the week to Oct 23, 2026. The leader here is Claude Code with 40T, so Letta Code scores 909B ÷ 40T × 100 = 2. Source
    • GitHub: 13,696 stars on github.com/letta-ai/letta-code on Oct 23, 2026. The leader here is OpenClaw with 393k, so Letta Code scores 3.7k ÷ 393k × 100 = 1. Source

    Not counted: VS Code, because it has no VS Code extension; JetBrains Developer Ecosystem 2026, because the survey didn't list it; Stack Overflow Developer Survey 2026, because the survey didn't list it.

    Usage score: (0 + 2 + 1) ÷ 3 = 1. Letta Code website

  18. 18
    Auggie
    Augment Code · closed source
    Usage score 0
    • npm 0 · 36k
    • VS Code 1 · 782k
    • npm: 035,740 downloads of @augmentcode/auggie in the 7 days to Nov 1, 2026. The leader here is Codex CLI with 28M, so Auggie scores 36k ÷ 28M × 100 = 0. Source
    • VS Code: 1782,491 installs of the augment.vscode-augment extension on Oct 23, 2026. The leader here is GitHub Copilot with 79M, so Auggie scores 782k ÷ 79M × 100 = 1. Source

    Not counted: OpenRouter, because it doesn't show up in OpenRouter's app rankings; GitHub, because it has no public GitHub repo; JetBrains Developer Ecosystem 2026, because the survey didn't list it; Stack Overflow Developer Survey 2026, because the survey didn't list it.

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

  19. 19
    Amp
    Sourcegraph · closed source
    Usage score 0
    • npm 0 · 26k
    • VS Code 0 · 108k
    • npm: 025,710 downloads of @sourcegraph/amp in the 7 days to Nov 1, 2026. The leader here is Codex CLI with 28M, so Amp scores 26k ÷ 28M × 100 = 0. Source
    • VS Code: 0108,457 installs of the sourcegraph.amp extension on Oct 23, 2026. The leader here is GitHub Copilot with 79M, so Amp scores 108k ÷ 79M × 100 = 0. Source

    Not counted: OpenRouter, because it doesn't show up in OpenRouter's app rankings; GitHub, because it has no public GitHub repo; JetBrains Developer Ecosystem 2026, because the survey didn't list it; Stack Overflow Developer Survey 2026, because the survey didn't list it.

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

Not ranked that day, only one source has them: Droid (npm, 33k).

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 Nov 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

  • OpenRouterTokens via OpenRouter, 7 days · data from Oct 23, 2026

    Tokens each app sent through OpenRouter in a week. It only sees traffic routed through OpenRouter, so agents that mostly call their own lab's API (Claude Code, Codex) look smaller than they are, and open-source agents that default to OpenRouter look bigger. Source

  • GitHubStars · data from Oct 23, 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 Oct 23, 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

  • JetBrains Developer Ecosystem 2026% of 15,000 respondents · data from Jul 31, 2026

    Share of developers in the survey who said they use each tool. Surveys only list the biggest tools. Source

  • Stack Overflow Developer Survey 2026% of 12,255 respondents · data from Oct 6, 2026

    Share of developers in the survey who said they use each tool. Surveys only list the biggest tools. 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 November 4, 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.