← Arena leaderboards

Vision · Style Control · Humor

Arena · 50 models

RankModelScore
1Claude Fable 5 (High)1,323.7309
2Muse Spark1,323.2087
3Gemini 3 Pro1,316.9875
4Gemini 3.1 Pro Preview1,299.9576
5Gemini 3.8 Flash (High)1,297.8712
6Claude Opus 4.6 (High)1,297.124
7Claude Opus 4.61,294.5187
8Gemini 3.5 Flash (High)1,292.8473
9Claude Fable 5.1 (Max)1,290.8672
10Claude Opus 4.7 (High)1,290.2887
11Qwen3.8-Max1,290.2259
12GPT-5.5 Instant1,286.9733
13Claude Opus 4.71,286.0493
14Gemini 3.5 Flash (Medium)1,285.6522
15GPT-5.5 (High)1,281.3138
16Gemini 3.5 Flash-Lite1,278.3703
17Gemini 3.7 Flash (High)1,275.3042
18Muse Spark 1.3 (xHigh)1,275.1707
19Muse Spark 1.11,274.9154
20Grok 4.6 (High)1,270.622
21Claude Opus 4.8 (High)1,269.7637
22Gemini 3 Flash1,268.7687
23Gemini 3.6 Flash (High)1,268.429
24Claude Opus 4.81,267.3763
25Muse Spark 1.3 (Max)1,267.0988
26GPT-5.51,265.6735
27Claude Opus 5 (High)1,265.3055
28GPT-5.11,264.2506
29Grok 4.20 Multi-Agent Beta (0309)1,263.7666
30GPT-5.4 (High)1,263.7306
31Grok 4.51,262.8609
32Kimi K2.5 (Thinking)1,262.1341
33Grok 4.20 Beta (Reasoning, 0309)1,261.7098
34Gemini 3 Flash (Minimal)1,261.7042
35Kimi K2.61,260.564
36Gemma 4 31B1,259.5837
37Dola Seed 2.0 Pro1,257.0837
38GPT-5.6 Sol (xHigh)1,256.6269
39GLM-5.3-Flash1,255.7024
40Gemini 3.1 Flash-Lite Preview1,250.2094
41Claude Sonnet 5 (High)1,248.8952
42Gemini 2.5 Pro1,247.1692
43Gemma 4 26B A4B1,245.2023
44GPT-5.41,245.0232
45DeepSeek-V4.1-Flash-20260910 (Max)1,243.3873
46Qwen3.5-397B-A17B1,241.9937
47Claude Sonnet 4.61,241.4159
48Qwen3.7-Plus1,238.9813
49GPT-5.2 Chat (20260210)1,237.5373
50GPT-5.4 mini (High)1,236.815