← Arena leaderboards

Text · Life & Physical & Social Science

Arena · 50 models

RankModelScore
1Gemini 4 Argon (High)1,546.4877
2Gemini 4 Carbon (High)1,542.7589
3Claude Fable 5.1 (Max)1,534.5701
4Claude Opus 5 (Max)1,532.3376
5Gemini 3.8 Flash (High)1,523.3633
6Claude Opus 5 (High)1,520.6079
7Claude Opus 5.5 (High)1,519.8126
8Claude Opus 4.6 (High)1,518.4528
9Qwen3.8-Max1,513.6275
10Claude Opus 4.61,511.391
11Gemini 3.7 Flash (High)1,511.281
12Kimi K3 (Max)1,507.7406
13Claude Fable 5 (High)1,507.5952
14Claude Opus 4.7 (High)1,507.5804
15MiMo-V2.6-Pro1,506.4613
16Muse Spark 1.3 (Max)1,506.0715
17Muse Wonder (Max)1,505.0773
18Muse Spark 1.2 (xHigh)1,499.1977
19GLM-5.3 (Max)1,498.0786
20Qwen3.5-Max-Preview1,497.5847
21Claude Opus 4.71,497.4711
22Gemini 3.1 Pro Preview1,496.8152
23Gemini 3.5 Flash (Medium)1,494.8679
24Gemini 3.5 Flash (High)1,493.8778
25Gemini 3.6 Flash (High)1,488.4907
26GLM-5.2 (Max)1,488.0554
27Claude Sonnet 5.5 (xHigh)1,486.9687
28GPT-5.5 (High)1,485.45
29Gemini 3 Pro1,484.9502
30GPT-5.51,484.5972
31GLM-5.3-Flash1,484.0522
32ERNIE 5.11,483.4707
33Muse Spark 1.11,483.1
34MiMo-V2.5-Pro1,481.7398
35Gemini 2.5 Pro1,480.063
36Qwen3.7-Max-Preview1,478.9854
37GLM-5.11,478.55
38GPT-5.4 (High)1,478.5038
39Claude Opus 4.8 (High)1,477.2174
40Muse Spark1,475.4451
41Claude Sonnet 4.61,474.5177
42Gemini 3 Flash1,472.5194
43Claude Opus 4.81,471.2784
44DeepSeek-V4.1-Flash-20260910 (Max)1,470.8104
45DeepSeek-V4-Pro1,470.0093
46ERNIE-5.0-Preview (1203)1,468.0704
47Kimi K2.61,467.835
48Qwen3.8-27B1,467.645
49Qwen3.6-Max-Preview1,466.0944
50Grok 4.51,464.5367