← Arena leaderboards

Text · Mathematical

Arena · 50 models

RankModelScore
1Gemini 4 Argon (High)1,548.8818
2Claude Fable 5.1 (Max)1,543.7894
3Claude Opus 5 (High)1,543.6848
4Claude Opus 5 (Max)1,539.1327
5Gemini 4 Carbon (High)1,534.1982
6Claude Opus 5.5 (High)1,526.4841
7Claude Opus 4.6 (High)1,524.9451
8Gemini 3.8 Flash (High)1,519.0891
9Claude Fable 5 (High)1,517.6078
10Claude Opus 4.61,515.8719
11GPT-6.1 Sol (Max)1,515.3743
12Claude Sonnet 5.5 (xHigh)1,513.3326
13MiMo-V2.6-Pro1,513.3171
14Claude Opus 4.7 (High)1,509.6875
15Kimi K3 (Max)1,504.0212
16Gemini 3.7 Flash (High)1,501.1947
17GLM-5.3 (Max)1,500.7573
18Muse Wonder (Max)1,499.3856
19Claude Opus 4.8 (High)1,496.8878
20GPT-5.4 (High)1,496.7603
21amazon-nova-experimental-chat (26-02-10)1,496.1718
22GPT-5.5 (High)1,495.9877
23Qwen3.8-Max1,495.9649
24GLM-5.3-Flash1,495.6477
25Claude Opus 4.71,495.3908
26Gemini 3.5 Flash (Medium)1,494.1423
27GPT-5.6 Sol (xHigh)1,493.7587
28Muse Spark 1.3 (Max)1,492.3481
29Gemini 3.5 Flash (High)1,492.2351
30ERNIE 5.11,491.4604
31Gemini 3.6 Flash (High)1,491.0855
32MiMo-V2.5-Pro1,491.0628
33GPT-5.51,490.0553
34Qwen3.5-Max-Preview1,490.0173
35GLM-5.2 (Max)1,489.3264
36Muse Spark 1.2 (xHigh)1,488.2983
37GPT-6 Astra (Max)1,486.6561
38Muse Spark 1.11,484.8613
39DeepSeek-V4.1-Flash-20260910 (Max)1,483.7082
40Hy31,483.4612
41Claude Sonnet 4.61,482.5725
42Claude Sonnet 5 (High)1,482.2816
43Grok 4.51,482.0096
44Claude Opus 4.81,481.1361
45Kimi K2.61,479.6697
46MiMo-V2.6-Flash1,478.6276
47Gemini 3.1 Pro Preview1,478.401
48GPT-5.6 Terra (xHigh)1,477.4343
49Gemma 4 26B A4B1,476.9994
50Beam 1.0 501B A23B1,476.1698