← Arena leaderboards

Vision · Homework

Arena · 50 models

RankModelScore
1Claude Opus 5.5 (High)1,362.0785
2Claude Fable 5 (High)1,360.8653
3GPT-6.1 Sol (Max)1,355.6054
4Gemini 3.8 Flash (High)1,351.4553
5Claude Fable 5.1 (Max)1,345.2299
6Claude Opus 5 (High)1,337.5687
7GPT-6 Astra (Max)1,337.4728
8Claude Opus 4.7 (High)1,337.3951
9Claude Opus 4.8 (High)1,336.8077
10Claude Opus 4.6 (High)1,336.7445
11GPT-5.4 (High)1,335.9297
12Qwen3.8-Max1,335.647
13Claude Opus 4.71,334.9114
14GPT-5.5 (High)1,331.7731
15Claude Sonnet 5.5 (xHigh)1,331.0949
16Gemini 3.5 Flash (Medium)1,329.8874
17Claude Opus 4.81,327.6181
18Claude Opus 4.61,326.6366
19GPT-5.51,326.6213
20Gemini 3.5 Flash (High)1,320.995
21Gemini 3.7 Flash (High)1,319.9898
22GLM-5.3-Flash1,317.7629
23Gemini 3.1 Pro Preview1,316.2455
24Gemini 3 Pro1,314.7347
25Gemini 3.6 Flash (High)1,313.7704
26Gemini 3 Flash1,312.5692
27GPT-5.6 Terra (xHigh)1,312.0972
28Claude Sonnet 4.61,311.2339
29Claude Sonnet 5 (High)1,311.0467
30Kimi K2.61,309.7278
31Gemma 4 31B1,309.7139
32GPT-5.41,309.2301
33Grok 4.7 (xHigh)1,309.022
34DeepSeek-V4.1-Flash-20260910 (Max)1,306.698
35GPT-5.6 Sol (xHigh)1,304.5141
36Muse Spark1,298.7851
37Gemini 3 Flash (Minimal)1,298.2304
38Kimi K2.5 (Thinking)1,296.6534
39Muse Spark 1.3 (xHigh)1,296.2007
40Muse Spark 1.3 (Max)1,295.5116
41Grok 4.51,295.4797
42Qwen3.7-Plus1,294.7267
43Qwen3.8-27B1,294.7086
44Gemma 4 26B A4B1,293.3058
45Qwen3.5-397B-A17B1,292.4249
46Gemini 3.5 Flash-Lite1,290.9056
47Muse Spark 1.11,286.9326
48GPT-6 Sol (Max)1,286.3464
49GPT-5.2 Chat (20260210)1,286.1644
50GPT-5.4 mini (High)1,285.4875