← Arena leaderboards

Text · Non-English

Arena · 50 models

RankModelScore
1Gemini 4 Carbon (High)1,536.9231
2Gemini 4 Argon (High)1,520.7306
3Claude Fable 5.1 (Max)1,509.6536
4Claude Opus 5.5 (High)1,505.953
5Claude Opus 5 (Max)1,501.5851
6Claude Opus 5 (High)1,497.5535
7Gemini 3.8 Flash (High)1,491.5002
8Muse Wonder (Max)1,490.2058
9Claude Opus 4.6 (High)1,487.5463
10Gemini 3.7 Flash (High)1,485.7317
11Claude Opus 4.61,483.8618
12Muse Spark 1.3 (Max)1,481.8758
13Claude Fable 5 (High)1,481.7599
14Claude Opus 4.7 (High)1,480.3278
15Gemini 3.1 Pro Preview1,476.2952
16Gemini 3.5 Flash (High)1,475.7214
17Muse Spark 1.2 (xHigh)1,475.1732
18MiMo-V2.6-Pro1,474.4305
19Claude Opus 4.71,474.3567
20Gemini 3 Pro1,474.2134
21Qwen3.7-Max-Preview1,472.9228
22Qwen3.8-Max1,472.389
23Gemini 3.5 Flash (Medium)1,471.9232
24Muse Spark 1.11,471.341
25Gemini 3.6 Flash (High)1,469.5366
26GPT-5.5 (High)1,467.1909
27Kimi K3 (Max)1,465.9284
28Qwen3.5-Max-Preview1,465.3788
29GPT-5.4 (High)1,465.1152
30GLM-5.3-Flash1,463.4363
31Muse Spark1,463.3407
32GPT-5.51,459.6546
33GLM-5.3 (Max)1,459.6391
34GLM-5.2 (Max)1,458.2322
35Gemini 3 Flash1,458.196
36Claude Sonnet 5.5 (xHigh)1,456.0958
37ERNIE 5.11,454.2112
38Gemini 2.5 Pro1,451.4373
39Claude Opus 4.8 (High)1,450.413
40GPT-5.6 Sol (xHigh)1,450.3564
41DeepSeek-V4.1-Flash-20260910 (Max)1,449.9126
42GLM-5.11,448.6472
43MiMo-V2.5-Pro1,448.4401
44GPT-5.41,448.0365
45Kimi K2.61,446.2633
46Qwen3.7-Plus1,444.6942
47Claude Opus 4.81,444.4355
48Dola Seed 2.0 Pro1,441.3059
49GPT-5.6 Terra (xHigh)1,441.013
50Grok 4.20 Beta (Reasoning, 0309)1,440.8563