ModelCompare
Compare

AI coding agent usage

What was the most used AI coding agent on ?

Claude Code from Anthropic, with a usage score of 65 out of 100 across 5 of 5 sources. GitHub Copilot was next at 53.

Usage score over time, week by week

Usage ranking

AI coding agents by usage on October 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. 5 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 65
    • npm 55 · 14M
    • OpenRouter 100 · 24T
    • GitHub 37 · 144k
    • VS Code 34 · 27M
    • JetBrains 100 · 39%
    • npm: 5514,471,471 downloads of @anthropic-ai/claude-code in the 7 days to Oct 1, 2026. The leader here is Codex CLI with 26M, so Claude Code scores 14M ÷ 26M × 100 = 55. Source
    • OpenRouter: 10024T tokens sent through OpenRouter by Claude Code in the week to Oct 1, 2026. That's the most of any agent in this source, so it scores 100. Source
    • GitHub: 37143,664 stars on github.com/anthropics/claude-code on Sep 1, 2026. The leader here is OpenClaw with 391k, so Claude Code scores 144k ÷ 391k × 100 = 37. Source
    • VS Code: 3426,755,687 installs of the anthropic.claude-code extension on Oct 1, 2026. The leader here is GitHub Copilot with 78M, so Claude Code scores 27M ÷ 78M × 100 = 34. 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

    Usage score: (55 + 100 + 37 + 34 + 100) ÷ 5 = 65. Claude Code website

  2. 2
    GitHub Copilot
    GitHub · closed source
    Usage score 53
    • npm 6 · 1.6M
    • VS Code 100 · 78M
    • JetBrains 54 · 21%
    • npm: 61,603,466 downloads of @github/copilot in the 7 days to Oct 1, 2026. The leader here is Codex CLI with 26M, so GitHub Copilot scores 1.6M ÷ 26M × 100 = 6. Source
    • VS Code: 10078,180,028 installs of the GitHub.copilot-chat extension on Sep 11, 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

    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) ÷ 3 = 53. GitHub Copilot website

  3. 3
    OpenClaw
    OpenClaw · open source
    Usage score 45
    • npm 14 · 3.7M
    • OpenRouter 22 · 5.4T
    • GitHub 100 · 391k
    • npm: 143,669,170 downloads of openclaw in the 7 days to Oct 1, 2026. The leader here is Codex CLI with 26M, so OpenClaw scores 3.7M ÷ 26M × 100 = 14. Source
    • OpenRouter: 225.4T tokens sent through OpenRouter by OpenClaw in the week to Oct 1, 2026. The leader here is Claude Code with 24T, so OpenClaw scores 5.4T ÷ 24T × 100 = 22. Source
    • GitHub: 100391,157 stars on github.com/openclaw/openclaw on Oct 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; JetBrains Developer Ecosystem 2026, because the survey didn't list it.

    Usage score: (14 + 22 + 100) ÷ 3 = 45. OpenClaw website

  4. 4
    Codex CLI
    OpenAI · open source
    Usage score 45
    • npm 100 · 26M
    • OpenRouter 33 · 8T
    • GitHub 33 · 127k
    • VS Code 18 · 14M
    • JetBrains 41 · 16%
    • npm: 10026,201,098 downloads of @openai/codex in the 7 days to Oct 1, 2026. That's the most of any agent in this source, so it scores 100. Source
    • OpenRouter: 338T tokens sent through OpenRouter by Codex CLI in the week to Oct 1, 2026. The leader here is Claude Code with 24T, so Codex CLI scores 8T ÷ 24T × 100 = 33. Source
    • GitHub: 33127,438 stars on github.com/openai/codex on Oct 1, 2026. The leader here is OpenClaw with 391k, so Codex CLI scores 127k ÷ 391k × 100 = 33. Source
    • VS Code: 1813,722,411 installs of the openai.chatgpt extension on Sep 1, 2026. The leader here is GitHub Copilot with 78M, so Codex CLI scores 14M ÷ 78M × 100 = 18. 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

    Usage score: (100 + 33 + 33 + 18 + 41) ÷ 5 = 45. Codex CLI website

  5. 5
    Cline
    Cline · open source
    Usage score 22
    • npm 0 · 95k
    • OpenRouter 64 · 16T
    • GitHub 18 · 70k
    • VS Code 7 · 5.2M
    • npm: 095,129 downloads of cline in the 7 days to Oct 1, 2026. The leader here is Codex CLI with 26M, so Cline scores 95k ÷ 26M × 100 = 0. Source
    • OpenRouter: 6416T tokens sent through OpenRouter by Cline in the week to Oct 1, 2026. The leader here is Claude Code with 24T, so Cline scores 16T ÷ 24T × 100 = 64. Source
    • GitHub: 1869,634 stars on github.com/cline/cline on Oct 1, 2026. The leader here is OpenClaw with 391k, so Cline scores 70k ÷ 391k × 100 = 18. Source
    • VS Code: 75,228,648 installs of the saoudrizwan.claude-dev extension on Sep 6, 2026. The leader here is GitHub Copilot with 78M, so Cline scores 5.2M ÷ 78M × 100 = 7. Source

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

    Usage score: (0 + 64 + 18 + 7) ÷ 4 = 22. Cline website

  6. 6
    OpenCode
    Anomaly · open source
    Usage score 22
    • npm 11 · 3M
    • GitHub 52 · 203k
    • VS Code 1 · 1.1M
    • npm: 112,989,468 downloads of opencode-ai in the 7 days to Oct 1, 2026. The leader here is Codex CLI with 26M, so OpenCode scores 3M ÷ 26M × 100 = 11. Source
    • GitHub: 52202,872 stars on github.com/anomalyco/opencode on Sep 1, 2026. The leader here is OpenClaw with 391k, so OpenCode scores 203k ÷ 391k × 100 = 52. Source
    • VS Code: 11,059,402 installs of the sst-dev.opencode extension on Sep 1, 2026. The leader here is GitHub Copilot with 78M, so OpenCode scores 1.1M ÷ 78M × 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.

    Usage score: (11 + 52 + 1) ÷ 3 = 22. OpenCode website

  7. 7
    Hermes Agent
    Nous Research · open source
    Usage score 21
    • OpenRouter 0 · 20B
    • GitHub 64 · 250k
    • OpenRouter: 020B tokens sent through OpenRouter by Hermes Agent in the week to Oct 1, 2026. The leader here is Claude Code with 24T, so Hermes Agent scores 20B ÷ 24T × 100 = 0. Source
    • GitHub: 64250,484 stars on github.com/NousResearch/hermes-agent on Oct 1, 2026. The leader here is OpenClaw with 391k, so Hermes Agent scores 250k ÷ 391k × 100 = 64. 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.

    Usage score: (0 + 64) ÷ 3 = 21, divided by 3 rather than 2 because every agent's average counts at least 3 sources. Hermes Agent website

  8. 8
    Kilo Code
    Kilo · open source
    Usage score 20
    • npm 0 · 47k
    • OpenRouter 69 · 17T
    • GitHub 7 · 27k
    • VS Code 2 · 1.5M
    • npm: 047,481 downloads of @kilocode/cli in the 7 days to Oct 1, 2026. The leader here is Codex CLI with 26M, so Kilo Code scores 47k ÷ 26M × 100 = 0. Source
    • OpenRouter: 6917T tokens sent through OpenRouter by Kilo Code in the week to Oct 1, 2026. The leader here is Claude Code with 24T, so Kilo Code scores 17T ÷ 24T × 100 = 69. Source
    • GitHub: 727,308 stars on github.com/Kilo-Org/kilocode on Sep 15, 2026. The leader here is OpenClaw with 391k, so Kilo Code scores 27k ÷ 391k × 100 = 7. Source
    • VS Code: 21,537,039 installs of the kilocode.Kilo-Code extension on Sep 15, 2026. The leader here is GitHub Copilot with 78M, so Kilo Code scores 1.5M ÷ 78M × 100 = 2. Source

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

    Usage score: (0 + 69 + 7 + 2) ÷ 4 = 20. Kilo Code website

  9. 9
    OpenHands
    All Hands AI · open source
    Usage score 14
    • OpenRouter 20 · 4.9T
    • GitHub 22 · 86k
    • OpenRouter: 204.9T tokens sent through OpenRouter by OpenHands in the week to Oct 1, 2026. The leader here is Claude Code with 24T, so OpenHands scores 4.9T ÷ 24T × 100 = 20. Source
    • GitHub: 2286,079 stars on github.com/OpenHands/OpenHands on Sep 3, 2026. The leader here is OpenClaw with 391k, so OpenHands scores 86k ÷ 391k × 100 = 22. 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.

    Usage score: (20 + 22) ÷ 3 = 14, divided by 3 rather than 2 because every agent's average counts at least 3 sources. OpenHands website

  10. 10
    pi
    Mario Zechner · open source
    Usage score 14
    • npm 0 · 7.6k
    • OpenRouter 41 · 10T
    • npm: 07,588 downloads of @mariozechner/pi-coding-agent in the 7 days to Oct 1, 2026. The leader here is Codex CLI with 26M, so pi scores 7.6k ÷ 26M × 100 = 0. Source
    • OpenRouter: 4110T tokens sent through OpenRouter by pi in the week to Oct 1, 2026. The leader here is Claude Code with 24T, so pi scores 10T ÷ 24T × 100 = 41. Source

    Not counted: GitHub, because there's no snapshot of its repo from around that day; VS Code, because it has no VS Code extension; JetBrains Developer Ecosystem 2026, because the survey didn't list it.

    Usage score: (0 + 41) ÷ 3 = 14, divided by 3 rather than 2 because every agent's average counts at least 3 sources. pi website

  11. 11
    Gemini CLI
    Google · open source
    Usage score 10
    • npm 2 · 454k
    • GitHub 27 · 107k
    • VS Code 1 · 1.1M
    • npm: 2454,428 downloads of @google/gemini-cli in the 7 days to Oct 1, 2026. The leader here is Codex CLI with 26M, so Gemini CLI scores 454k ÷ 26M × 100 = 2. Source
    • GitHub: 27106,823 stars on github.com/google-gemini/gemini-cli on Sep 5, 2026. The leader here is OpenClaw with 391k, so Gemini CLI scores 107k ÷ 391k × 100 = 27. Source
    • VS Code: 11,079,995 installs of the Google.gemini-cli-vscode-ide-companion extension on Sep 16, 2026. The leader here is GitHub Copilot with 78M, so Gemini CLI scores 1.1M ÷ 78M × 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.

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

  12. 12
    Aider
    Aider · open source
    Usage score 4
    • OpenRouter 0 · 0
    • GitHub 12 · 49k
    • OpenRouter: 00 tokens sent through OpenRouter by Aider in the week to Oct 1, 2026. The leader here is Claude Code with 24T, so Aider scores 0 ÷ 24T × 100 = 0. Source
    • GitHub: 1248,648 stars on github.com/Aider-AI/aider on Sep 1, 2026. The leader here is OpenClaw with 391k, so Aider scores 49k ÷ 391k × 100 = 12. 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.

    Usage score: (0 + 12) ÷ 3 = 4, divided by 3 rather than 2 because every agent's average counts at least 3 sources. Aider website

  13. 13
    Roo Code
    Roo Code · open source
    Usage score 3
    • OpenRouter 1 · 219B
    • GitHub 6 · 24k
    • VS Code 2 · 1.9M
    • OpenRouter: 1219B tokens sent through OpenRouter by Roo Code in the week to Oct 1, 2026. The leader here is Claude Code with 24T, so Roo Code scores 219B ÷ 24T × 100 = 1. Source
    • GitHub: 624,337 stars on github.com/RooCodeInc/Roo-Code on Aug 16, 2026. The leader here is OpenClaw with 391k, so Roo Code scores 24k ÷ 391k × 100 = 6. Source
    • VS Code: 21,940,604 installs of the RooVeterinaryInc.roo-cline extension on Aug 20, 2026. The leader here is GitHub Copilot with 78M, so Roo Code scores 1.9M ÷ 78M × 100 = 2. Source

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

    Usage score: (1 + 6 + 2) ÷ 3 = 3. Roo Code website

  14. 14
    Qwen Code
    Alibaba · open source
    Usage score 3
    • npm 0 · 92k
    • OpenRouter 1 · 180B
    • GitHub 7 · 28k
    • npm: 092,247 downloads of @qwen-code/qwen-code in the 7 days to Oct 1, 2026. The leader here is Codex CLI with 26M, so Qwen Code scores 92k ÷ 26M × 100 = 0. Source
    • OpenRouter: 1180B tokens sent through OpenRouter by Qwen Code in the week to Oct 1, 2026. The leader here is Claude Code with 24T, so Qwen Code scores 180B ÷ 24T × 100 = 1. Source
    • GitHub: 728,270 stars on github.com/QwenLM/qwen-code on Oct 2, 2026. The leader here is OpenClaw with 391k, so Qwen Code scores 28k ÷ 391k × 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.

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

  15. 15
    Crush
    Charm · open source
    Usage score 2
    • npm 0 · 8.7k
    • GitHub 7 · 28k
    • npm: 08,749 downloads of @charmland/crush in the 7 days to Oct 1, 2026. The leader here is Codex CLI with 26M, so Crush scores 8.7k ÷ 26M × 100 = 0. Source
    • GitHub: 727,873 stars on github.com/charmbracelet/crush on Sep 3, 2026. The leader here is OpenClaw with 391k, so Crush scores 28k ÷ 391k × 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.

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

  16. 16
    Letta Code
    Letta · open source
    Usage score 1
    • npm 1 · 213k
    • OpenRouter 2 · 387B
    • GitHub 1 · 3.3k
    • npm: 1212,667 downloads of @letta-ai/letta-code in the 7 days to Oct 1, 2026. The leader here is Codex CLI with 26M, so Letta Code scores 213k ÷ 26M × 100 = 1. Source
    • OpenRouter: 2387B tokens sent through OpenRouter by Letta Code in the week to Oct 1, 2026. The leader here is Claude Code with 24T, so Letta Code scores 387B ÷ 24T × 100 = 2. Source
    • GitHub: 13,333 stars on github.com/letta-ai/letta-code on Sep 14, 2026. The leader here is OpenClaw with 391k, so Letta Code scores 3.3k ÷ 391k × 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.

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

  17. 17
    Auggie
    Augment Code · closed source
    Usage score 0
    • npm 0 · 38k
    • VS Code 1 · 778k
    • npm: 037,723 downloads of @augmentcode/auggie in the 7 days to Oct 1, 2026. The leader here is Codex CLI with 26M, so Auggie scores 38k ÷ 26M × 100 = 0. Source
    • VS Code: 1777,579 installs of the augment.vscode-augment extension on Sep 15, 2026. The leader here is GitHub Copilot with 78M, so Auggie scores 778k ÷ 78M × 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.

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

Not ranked that day, only one source has them: Cursor (JetBrains, 12%), Amp (npm, 29k), Droid (npm, 10k).

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 Oct 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 1, 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 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 Oct 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

  • 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

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.