← Arena leaderboards

Text · Style Control

Arena · 50 models

RankModelScore
1Gemini 4 Carbon (High)1,527.5065
2Gemini 4 Argon (High)1,520.2546
3Claude Opus 5.5 (High)1,508.6976
4Muse Wonder (Max)1,507.0977
5Claude Fable 5.1 (Max)1,504.3663
6Claude Fable 5 (High)1,503.8924
7Claude Opus 4.6 (High)1,503.1628
8Claude Opus 4.7 (High)1,501.0498
9Claude Opus 4.61,496.0961
10Gemini 3.8 Flash (High)1,495.9451
11Muse Spark 1.3 (Max)1,495.0456
12Claude Opus 4.71,493.7218
13Muse Spark 1.2 (xHigh)1,492.188
14Claude Opus 5 (High)1,491.5506
15Claude Opus 5 (Max)1,490.504
16Muse Spark 1.11,490.2352
17Gemini 3.7 Flash (High)1,489.5501
18Muse Spark1,489.0251
19Kimi K3 (Max)1,488.1895
20Gemini 3.1 Pro Preview1,486.9
21Gemini 3 Pro1,485.4329
22GPT-5.6 Sol (xHigh)1,484.9105
23Gemini 3.6 Flash (High)1,483.1551
24Qwen3.8-Max1,482.4121
25Claude Opus 4.8 (High)1,481.7111
26GPT-5.5 (High)1,481.6951
27GPT-6.1 Sol (Max)1,481.328
28GLM-5.3 (Max)1,480.5694
29MiMo-V2.6-Pro1,479.6979
30Gemini 3.5 Flash (High)1,477.6274
31Claude Sonnet 5.5 (xHigh)1,477.1495
32GPT-5.51,476.6502
33GLM-5.3-Flash1,476.5156
34Gemini 3.5 Flash (Medium)1,476.4084
35DeepSeek-V4.1-Flash-20260910 (Max)1,476.2276
36GPT-5.2 Chat (20260210)1,475.6639
37GPT-6 Astra (Max)1,475.6479
38GLM-5.2 (Max)1,475.2898
39GPT-5.4 (High)1,475.2742
40Grok 4.20 Beta 11,474.6133
41Claude Opus 4.81,474.4404
42Qwen3.7-Max-Preview1,473.7034
43Claude Opus 4.5 (High, 32K, 20251101)1,473.6905
44GPT-5.5 Instant1,473.1666
45Gemini 3 Flash1,472.7476
46Claude Sonnet 4.61,471.9507
47Grok 4.20 Beta (Reasoning, 0309)1,471.74
48Grok 4.20 Multi-Agent Beta (0309)1,470.7061
49Claude Opus 4.5 (20251101)1,469.6756
50GPT-5.6 Terra (xHigh)1,467.3373