WebDev Arena (Frontend): leaderboard
Voters submit a prompt for a web app, two anonymous models build it, and both builds run in sandboxes the voter can click through before choosing one. This board is the frontend slice, where models generate HTML or React front ends, as opposed to the fullstack slice, where prompts call for a database, API, authentication or other backend work. Votes are fitted with a Bradley-Terry model and higher is better.
Metric: Arena Score. Source: arena.ai. 142 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5.5 (Max) | 1857 | #3 (Claude Opus 5.5) |
| 2 | Claude Sonnet 5.5 (xHigh) | 1813 | #11 (Claude Sonnet 5.5) |
| 3 | GPT-6 Astra (Max) | 1809 | #6 (GPT-6 Astra) |
| 4 | Claude Fable 5.1 (Max) | 1778 | #9 (Claude Fable 5.1) |
| 5 | GPT-6.1 Sol (Max) | 1771 | #7 (GPT-6.1 Sol) |
| 6 | Claude Sonnet 5.5 (High) | 1745 | #11 (Claude Sonnet 5.5) |
| 7 | Claude Opus 5 (Max) | 1713 | #12 (Claude Opus 5) |
| 8 | GPT-6 Sol (Max) | 1695 | #14 (GPT-6 Sol) |
| 9 | Gemini 4 Argon (High) | 1687 | #8 (Gemini 4 Argon) |
| 10 | Qwen 3.8 Max (0902) | 1686 | #29 |
| 11 | Qwen 3.8 Max | 1679 | #35 |
| 12 | Claude Opus 5 (High) | 1677 | #12 (Claude Opus 5) |
| 13 | Muse Spark 1.3 (Max) | 1672 | #24 (Muse Spark 1.3) |
| 14 | Kimi K3 (Max) | 1669 | #30 (Kimi K3) |
| 15 | Grok 4.7 (xHigh) | 1658 | #25 (Grok 4.7) |
Interactive version: theaggregate.ai/benchmark?slug=webdev-arena-frontend · How It Works · Data refreshed daily, snapshot 2026-10-20.