← Arena leaderboards

WebDev · Image-to-WebDev

Arena · 50 models

RankModelScore
1Claude Opus 5.5 (Max)1,752.5704
2Claude Sonnet 5.5 (xHigh)1,737.1398
3GPT-6 Astra (Max)1,726.2123
4Claude Fable 5.1 (Max)1,722.2191
5GPT-6.1 Sol (Max)1,701.7353
6Claude Opus 5 (Max)1,661.8189
7GPT-6 Sol (Max)1,660.4796
8Muse Spark 1.3 (Max)1,640.3894
9Qwen3.8-Max (0902)1,636.5884
10Claude Fable 5 (High)1,622.8328
11Qwen3.8-Max1,614.4987
12GPT-5.6 Sol (xHigh, Codex Harness)1,609.166
13DeepSeek-V4.1-Flash-20260910 (Max)1,606.33
14Grok 4.6 (High)1,603.4686
15Gemini 3.7 Flash (High)1,599.488
16MiMo-V2.6-Pro1,598.2806
17Muse Spark 1.3 (xHigh)1,595.9168
18MiMo-V2.6-Flash1,595.5408
19Gemini 3.8 Flash (High)1,592.7182
20Grok 4.7 (xHigh)1,585.5229
21Kimi K3 (Max)1,578.7772
22GLM-5.3-Flash1,578.3233
23Claude Opus 4.7 (High)1,577.4721
24Qwen3.8-27B1,577.4376
25GPT-6 Luna (Max)1,571.1091
26Claude Opus 4.71,561.8289
27Grok 4.51,561.681
28Step 5 Preview (High)1,550.4117
29Gemini 3.6 Flash (High)1,545.4402
30Claude Opus 4.6 (High)1,542.7388
31GPT-5.6 Terra (xHigh, Codex Harness)1,540.5135
32Claude Sonnet 5 (High)1,539.862
33Muse Spark 1.11,539.58
34Claude Opus 4.61,538.3272
35GPT-5.5 (xHigh, Codex Harness)1,532.086
36Claude Sonnet 4.61,529.4437
37Kimi K2.61,520.8565
38GPT-5.5 (High, Codex Harness)1,519.3783
39GPT-5.6 Luna (xHigh, Codex Harness)1,516.9857
40Kimi K2.7 Code1,512.8899
41Seed2.1 Pro Preview1,509.5136
42Gemini 3.5 Flash (Medium)1,504.3705
43GPT-5.5 (Codex Harness)1,503.218
44Gemini 3.5 Flash (High)1,501.1895
45Gemini 3.1 Pro Preview1,481.0808
46Qwen3.6-Plus1,472.0853
47MiniMax-M31,470.3442
48Gemini 3 Flash1,450.2681
49Gemini 3.5 Flash-Lite1,447.7004
50GPT-5.3-Codex (Codex Harness)1,443.1393