← Arena leaderboards

WebDev · HTML

Arena · 50 models

RankModelScore
1GPT-6.1 Sol (Max)1,795.8032
2GPT-6 Astra (Max)1,784.8782
3Gemini 4 Carbon (High)1,779.8253
4Claude Opus 5.5 (Max)1,753.389
5Claude Sonnet 5.5 (xHigh)1,738.6281
6Muse Wonder Full (Max)1,730.9552
7GPT-6.1 Terra (Max)1,718.763
8Kimi K3.1 Pro (Max)1,706.1363
9DeepSeek-V4.1-Flash-20261017 (Max)1,700.8664
10Claude Sonnet 5.5 (High)1,699.8566
11Claude Fable 5.1 (Max)1,682.8046
12GPT-6 Sol (Max)1,679.628
13Claude Opus 5 (Max)1,677.0826
14Muse Wonder (Max)1,669.637
15Gemini 4 Argon (High)1,667.0403
16Qwen3.8-Max (0902)1,657.3836
17Muse Spark 1.3 (Max)1,654.4242
18Claude Fable 5 (High)1,650.6759
19Claude Opus 5 (High)1,649.0753
20Kimi K3 (Max)1,640.9694
21MiMo-V2.6-Flash1,638.4337
22Grok 4.7 (xHigh)1,628.5726
23Qwen3.8-Max1,625.924
24MiMo-V2.6-Pro1,624.2985
25Hy4 preview1,620.0388
26Muse Spark 1.3 (xHigh)1,619.8695
27GPT-5.6 Sol (xHigh, Codex Harness)1,617.0472
28DeepSeek-V4.1-Flash-20260910 (Max)1,613.0479
29Gemini 3.8 Flash (High)1,612.4468
30Grok 4.6 (High)1,597.0297
31GLM-5.3 (Max)1,596.9038
32Qwen3.8-Flash-Next1,594.8998
33GPT-6 Luna (Max)1,594.8985
34Gemini 3.7 Flash (High)1,590.6633
35GLM-5.3-Flash1,577.1371
36Claude Haiku 5.5 (High)1,567.9554
37GLM-5.2 (Max)1,567.6147
38Grok 4.51,567.5525
39DeepSeek-V4-Pro (High, 20260813)1,565.4488
40Qwen3.8-27B1,565.0953
41Claude Opus 4.6 (High)1,558.6412
42DeepSeek-V4-Flash (High)1,557.5726
43Step 5 Preview (High)1,555.0615
44Claude Opus 4.71,554.3731
45Gemini 3.6 Flash (High)1,546.5416
46Claude Opus 4.61,546.2211
47Claude Opus 4.7 (High)1,546.0919
48Muse Spark 1.2 (xHigh)1,543.0411
49Claude Opus 4.8 (High)1,542.7204
50GPT-5.5 (xHigh, Codex Harness)1,542.6578