WebDev Arena (Gaming): leaderboard

arena.ai’s WebDev Arena restricted to browser game builds: two models answer the same web-app prompt, the rendered results are shown side by side, and the votes are fitted to an Arena Score over that category alone.

Metric: Arena Score. Source: arena.ai. 141 models tracked.

Top models

#ModelScoreOverall rank
1Claude Opus 5.5 (Max)1935#3 (Claude Opus 5.5)
2Claude Sonnet 5.5 (xHigh)1892#14 (Claude Sonnet 5.5)
3GPT-6 Astra (Max)1872#6 (GPT-6 Astra)
4Claude Fable 5.1 (Max)1858#10 (Claude Fable 5.1)
5GPT-6.1 Sol (Max)1839#7 (GPT-6.1 Sol)
6Claude Sonnet 5.5 (High)1833#14 (Claude Sonnet 5.5)
7Gemini 4 Argon (High)1768#9 (Gemini 4 Argon)
8Claude Opus 5 (Max)1766#16 (Claude Opus 5)
9GPT-6 Sol (Max)1757#19 (GPT-6 Sol)
10Qwen 3.8 Max (0902)1740#36
11Qwen 3.8 Max1737#34
12Claude Opus 5 (High)1727#16 (Claude Opus 5)
13Muse Spark 1.3 (Max)1726#31 (Muse Spark 1.3)
14MiMo-V2.6-Flash1708#74
15Claude Haiku 5.5 (High)1706#43 (Claude Haiku 5.5)

Interactive version: theaggregate.ai/benchmark?slug=webdev-arena-gaming · How It Works · Data refreshed daily, snapshot 2026-11-05.