ModelCompare
Compare

AI coding agent usage

What was the most used AI coding agent on ?

Claude Code from Anthropic, with a usage score of 48 out of 100 across 3 of 3 sources. Gemini CLI was next at 35.

Usage score over time, week by week

Usage ranking

AI coding agents by usage on September 1, 2025

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 48
    • npm 100 · 5.2M
    • GitHub 44 · 32k
    • VS Code 1 · 452k
    • npm: 1005,246,852 downloads of @anthropic-ai/claude-code in the 7 days to Sep 1, 2025. That's the most of any agent in this source, so it scores 100. Source
    • GitHub: 4432,130 stars on github.com/anthropics/claude-code on Sep 1, 2025. The leader here is Gemini CLI with 73k, so Claude Code scores 32k ÷ 73k × 100 = 44. Source
    • VS Code: 1451,936 installs of the anthropic.claude-code extension on Aug 16, 2025. The leader here is GitHub Copilot with 37M, so Claude Code scores 452k ÷ 37M × 100 = 1. Source

    Usage score: (100 + 44 + 1) ÷ 3 = 48. Claude Code website

  2. 2
    Gemini CLI
    Google · open source
    Usage score 35
    • npm 5 · 273k
    • GitHub 100 · 73k
    • VS Code 0 · 6.4k
    • npm: 5273,261 downloads of @google/gemini-cli in the 7 days to Sep 1, 2025. The leader here is Claude Code with 5.2M, so Gemini CLI scores 273k ÷ 5.2M × 100 = 5. Source
    • GitHub: 10072,976 stars on github.com/google-gemini/gemini-cli on Sep 1, 2025. That's the most of any agent in this source, so it scores 100. Source
    • VS Code: 06,397 installs of the Google.gemini-cli-vscode-ide-companion extension on Aug 14, 2025. The leader here is GitHub Copilot with 37M, so Gemini CLI scores 6.4k ÷ 37M × 100 = 0. Source

    Usage score: (5 + 100 + 0) ÷ 3 = 35. Gemini CLI website

  3. 3
    GitHub Copilot
    GitHub · closed source
    Usage score 33
    • npm 0 · 0
    • VS Code 100 · 37M
    • npm: 00 downloads of @github/copilot in the 7 days to Sep 1, 2025. The leader here is Claude Code with 5.2M, so GitHub Copilot scores 0 ÷ 5.2M × 100 = 0. Source
    • VS Code: 10036,932,593 installs of the GitHub.copilot-chat extension on Aug 11, 2025. 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: (0 + 100) ÷ 3 = 33, divided by 3 rather than 2 because every agent's average counts at least 3 sources. GitHub Copilot website

  4. 4
    Cline
    Cline · open source
    Usage score 25
    • npm 0 · 2.9k
    • GitHub 68 · 50k
    • VS Code 6 · 2M
    • npm: 02,941 downloads of cline in the 7 days to Sep 1, 2025. The leader here is Claude Code with 5.2M, so Cline scores 2.9k ÷ 5.2M × 100 = 0. Source
    • GitHub: 6849,951 stars on github.com/cline/cline on Sep 1, 2025. The leader here is Gemini CLI with 73k, so Cline scores 50k ÷ 73k × 100 = 68. Source
    • VS Code: 62,045,044 installs of the saoudrizwan.claude-dev extension on Aug 11, 2025. The leader here is GitHub Copilot with 37M, so Cline scores 2M ÷ 37M × 100 = 6. Source

    Usage score: (0 + 68 + 6) ÷ 3 = 25. Cline website

  5. 5
    Codex CLI
    OpenAI · open source
    Usage score 16
    • npm 3 · 150k
    • GitHub 43 · 32k
    • VS Code 1 · 423k
    • npm: 3149,689 downloads of @openai/codex in the 7 days to Sep 1, 2025. The leader here is Claude Code with 5.2M, so Codex CLI scores 150k ÷ 5.2M × 100 = 3. Source
    • GitHub: 4331,688 stars on github.com/openai/codex on Aug 2, 2025. The leader here is Gemini CLI with 73k, so Codex CLI scores 32k ÷ 73k × 100 = 43. Source
    • VS Code: 1423,490 installs of the openai.chatgpt extension on Aug 8, 2025. The leader here is GitHub Copilot with 37M, so Codex CLI scores 423k ÷ 37M × 100 = 1. Source

    Usage score: (3 + 43 + 1) ÷ 3 = 16. Codex CLI website

  6. 6
    Roo Code
    Roo Code · open source
    Usage score 9
    • GitHub 25 · 18k
    • VS Code 2 · 643k
    • GitHub: 2517,927 stars on github.com/RooCodeInc/Roo-Code on Aug 3, 2025. The leader here is Gemini CLI with 73k, so Roo Code scores 18k ÷ 73k × 100 = 25. Source
    • VS Code: 2642,661 installs of the RooVeterinaryInc.roo-cline extension on Jul 9, 2025. The leader here is GitHub Copilot with 37M, so Roo Code scores 643k ÷ 37M × 100 = 2. Source

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

    Usage score: (25 + 2) ÷ 3 = 9, divided by 3 rather than 2 because every agent's average counts at least 3 sources. Roo Code website

  7. 7
    Qwen Code
    Alibaba · open source
    Usage score 6
    • npm 1 · 32k
    • GitHub 16 · 12k
    • npm: 132,398 downloads of @qwen-code/qwen-code in the 7 days to Sep 1, 2025. The leader here is Claude Code with 5.2M, so Qwen Code scores 32k ÷ 5.2M × 100 = 1. Source
    • GitHub: 1611,837 stars on github.com/QwenLM/qwen-code on Sep 1, 2025. The leader here is Gemini CLI with 73k, so Qwen Code scores 12k ÷ 73k × 100 = 16. Source

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

    Usage score: (1 + 16) ÷ 3 = 6, divided by 3 rather than 2 because every agent's average counts at least 3 sources. Qwen Code website

  8. 8
    Kilo Code
    Kilo · open source
    Usage score 3
    • npm 0 · 0
    • GitHub 9 · 6.4k
    • VS Code 1 · 190k
    • npm: 00 downloads of @kilocode/cli in the 7 days to Sep 1, 2025. The leader here is Claude Code with 5.2M, so Kilo Code scores 0 ÷ 5.2M × 100 = 0. Source
    • GitHub: 96,436 stars on github.com/Kilo-Org/kilocode on Aug 7, 2025. The leader here is Gemini CLI with 73k, so Kilo Code scores 6.4k ÷ 73k × 100 = 9. Source
    • VS Code: 1189,594 installs of the kilocode.Kilo-Code extension on Aug 8, 2025. The leader here is GitHub Copilot with 37M, so Kilo Code scores 190k ÷ 37M × 100 = 1. Source

    Usage score: (0 + 9 + 1) ÷ 3 = 3. Kilo Code website

  9. 9
    Crush
    Charm · open source
    Usage score 1
    • npm 0 · 2k
    • GitHub 4 · 3.2k
    • npm: 02,021 downloads of @charmland/crush in the 7 days to Sep 1, 2025. The leader here is Claude Code with 5.2M, so Crush scores 2k ÷ 5.2M × 100 = 0. Source
    • GitHub: 43,228 stars on github.com/charmbracelet/crush on Aug 1, 2025. The leader here is Gemini CLI with 73k, so Crush scores 3.2k ÷ 73k × 100 = 4. Source

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

    Usage score: (0 + 4) ÷ 3 = 1, divided by 3 rather than 2 because every agent's average counts at least 3 sources. Crush website

  10. 10
    Auggie
    Augment Code · closed source
    Usage score 0
    • npm 0 · 8.4k
    • VS Code 1 · 411k
    • npm: 08,431 downloads of @augmentcode/auggie in the 7 days to Sep 1, 2025. The leader here is Claude Code with 5.2M, so Auggie scores 8.4k ÷ 5.2M × 100 = 0. Source
    • VS Code: 1410,986 installs of the augment.vscode-augment extension on Aug 1, 2025. The leader here is GitHub Copilot with 37M, so Auggie scores 411k ÷ 37M × 100 = 1. Source

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

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

  11. 11
    Amp
    Sourcegraph · closed source
    Usage score 0
    • npm 0 · 15k
    • VS Code 0 · 16k
    • npm: 015,404 downloads of @sourcegraph/amp in the 7 days to Sep 1, 2025. The leader here is Claude Code with 5.2M, so Amp scores 15k ÷ 5.2M × 100 = 0. Source
    • VS Code: 015,623 installs of the sourcegraph.amp extension on Jul 9, 2025. The leader here is GitHub Copilot with 37M, so Amp scores 16k ÷ 37M × 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

  12. 12
    OpenCode
    Anomaly · open source
    Usage score 0
    • npm 0 · 12k
    • VS Code 0 · 1.8k
    • npm: 011,833 downloads of opencode-ai in the 7 days to Sep 1, 2025. The leader here is Claude Code with 5.2M, so OpenCode scores 12k ÷ 5.2M × 100 = 0. Source
    • VS Code: 01,847 installs of the sst-dev.opencode extension on Aug 13, 2025. The leader here is GitHub Copilot with 37M, so OpenCode scores 1.8k ÷ 37M × 100 = 0. Source

    Not counted: GitHub, because there's no snapshot of its repo from around that day.

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

Not ranked that day, only one source has them: Aider (GitHub, 36k), Droid (npm, 80).

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 Sep 1, 2025

    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 Sep 1, 2025

    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 Aug 16, 2025

    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.