← Arena leaderboards

Text · Style Control · Non-English

Arena · 50 models

RankModelScore
1Gemini 4 Carbon (High)1,520.763
2Gemini 4 Argon (High)1,510.0306
3Claude Fable 5.1 (Max)1,497.3863
4Claude Opus 5.5 (High)1,497.0043
5Muse Wonder (Max)1,496.9151
6Claude Fable 5 (High)1,492.9015
7Claude Opus 4.7 (High)1,489.5536
8Claude Opus 4.6 (High)1,488.4724
9Gemini 3.8 Flash (High)1,486.9119
10Gemini 3.7 Flash (High)1,484.7815
11Muse Spark 1.2 (xHigh)1,484.6957
12Muse Spark 1.3 (Max)1,484.4685
13Claude Opus 4.61,483.0317
14Claude Opus 4.71,482.9604
15Muse Spark 1.11,482.9337
16Claude Opus 5 (High)1,481.9635
17Claude Opus 5 (Max)1,481.5803
18Gemini 3.1 Pro Preview1,479.0556
19Gemini 3 Pro1,477.3795
20Muse Spark1,477.1546
21Kimi K3 (Max)1,477.0979
22GPT-5.6 Sol (xHigh)1,475.2058
23GPT-5.5 (High)1,473.911
24Gemini 3.6 Flash (High)1,471.4099
25Qwen3.8-Max1,471.1806
26Claude Opus 4.8 (High)1,469.5811
27GPT-5.5 Instant1,469.3594
28Qwen3.7-Max-Preview1,469.1159
29Gemini 3.5 Flash (High)1,468.9806
30GPT-6.1 Sol (Max)1,468.347
31GPT-5.4 (High)1,468.1847
32Gemini 3.5 Flash (Medium)1,467.4384
33GPT-5.51,466.21
34GLM-5.3-Flash1,465.5842
35GLM-5.3 (Max)1,465.4468
36Gemini 3 Flash1,463.8006
37Claude Opus 4.81,463.7758
38GPT-5.2 Chat (20260210)1,463.1453
39Grok 4.20 Beta 11,463.1166
40GPT-6 Astra (Max)1,462.5946
41GLM-5.2 (Max)1,461.7612
42Grok 4.20 Beta (Reasoning, 0309)1,461.3736
43DeepSeek-V4.1-Flash-20260910 (Max)1,460.658
44MiMo-V2.6-Pro1,460.3757
45Grok 4.20 Multi-Agent Beta (0309)1,460.2452
46Claude Opus 4.5 (High, 32K, 20251101)1,459.9707
47Claude Sonnet 5.5 (xHigh)1,458.3694
48Claude Sonnet 4.61,457.1116
49Claude Opus 4.5 (20251101)1,456.724
50GPT-5.41,456.7102