Atlas

Benchmarks

← All benchmarks

Arena Code WebDev — React

Code

Arena Code WebDev rating for websites and applications generated with React. Anonymous outputs are compared through blind pairwise user votes.

Top models (higher is better)

ModelScore
Claude Opus 5.51856
Claude Sonnet 5.51816
GPT-6 Astra1796
Claude Fable 5.11791
GPT-6.1 Sol1754
Muse Wonder Full1740
Claude Opus 51713
Kimi K3.1 Pro1697
Gemini 4 Argon1690
Qwen3.8 Max (0902)1689
GPT-6 Sol1684
Qwen3.8-Max1683
Muse Spark 1.31671
Kimi K31670
DeepSeek V4.1 Flash 10171666
Grok 4.71665
Qwen3.8-Flash-Next1655
Hy4 Preview1653
MiMo-V2.6-Flash1650
Muse Wonder1645
GLM-5.31643
MiMo-V2.6-Pro1634
Claude Fable 51631
Grok 4.61630
DeepSeek V4.1 Flash1628
GLM-5.3 Flash1625
GPT-5.6 Sol1619
GLM-5.21617
MiniMax M3.1 Flash1612
Claude Haiku 5.51606
Qwen3.8 27B1600
DeepSeek V4 Pro 08131588
Gemini 3.7 Flash1587
DeepSeek-V4-Flash1585
Gemini 3.8 Flash1577
GPT-6 Luna1574
Opus 4.71560
Mistral Large 4 Preview1560
Opus 4.81557
Grok 4.51556
Loading Atlas data…