← Arena leaderboards

Text

Arena · 50 models

RankModelScore
1Gemini 4 Carbon (High)1,540.0421
2Gemini 4 Argon (High)1,528.577
3Claude Opus 5.5 (High)1,515.7998
4Claude Fable 5.1 (Max)1,513.8339
5Claude Opus 5 (Max)1,507.9091
6Claude Opus 5 (High)1,504.0843
7Claude Opus 4.6 (High)1,501.7236
8Muse Wonder (Max)1,499.1666
9Gemini 3.8 Flash (High)1,498.4923
10Claude Opus 4.61,496.2496
11Claude Fable 5 (High)1,491.5611
12MiMo-V2.6-Pro1,491.3511
13Muse Spark 1.3 (Max)1,490.6256
14Claude Opus 4.7 (High)1,489.7196
15Gemini 3.7 Flash (High)1,489.1174
16Qwen3.8-Max1,483.2906
17Claude Opus 4.71,482.9868
18Muse Spark 1.2 (xHigh)1,482.6648
19Gemini 3.5 Flash (High)1,482.1753
20Gemini 3.1 Pro Preview1,480.1062
21Gemini 3.6 Flash (High)1,479.7907
22Gemini 3 Pro1,479.4014
23Muse Spark 1.11,478.2816
24Gemini 3.5 Flash (Medium)1,478.1611
25Kimi K3 (Max)1,475.6886
26Qwen3.7-Max-Preview1,474.6659
27Muse Spark1,474.0352
28GLM-5.3 (Max)1,473.7757
29GLM-5.3-Flash1,473.0372
30Claude Sonnet 5.5 (xHigh)1,472.8511
31GPT-5.5 (High)1,471.5687
32Qwen3.5-Max-Preview1,470.3945
33GLM-5.2 (Max)1,469.6434
34GPT-5.4 (High)1,469.0796
35ERNIE 5.11,467.7697
36GPT-5.51,466.752
37Gemini 3 Flash1,465.6661
38DeepSeek-V4.1-Flash-20260910 (Max)1,464.507
39MiMo-V2.5-Pro1,464.2246
40GLM-5.11,462.8156
41Claude Opus 4.8 (High)1,461.2951
42Claude Sonnet 4.61,457.7072
43Gemini 2.5 Pro1,457.701
44GPT-5.6 Sol (xHigh)1,457.2397
45Kimi K2.61,455.557
46Qwen3.7-Plus1,454.5577
47Claude Opus 4.81,454.0343
48GPT-5.41,452.4698
49MiMo-V2.6-Flash1,451.2184
50DeepSeek-V4-Pro1,451.1279