← Arena leaderboards

Text · Style Control · Life & Physical & Social Science

Arena · 50 models

RankModelScore
1Gemini 4 Argon (High)1,536.3095
2Gemini 4 Carbon (High)1,530.1969
3Claude Opus 4.6 (High)1,526.8875
4Claude Fable 5 (High)1,526.5527
5Claude Opus 4.7 (High)1,525.0926
6Kimi K3 (Max)1,523.7742
7Claude Fable 5.1 (Max)1,522.9507
8Muse Wonder (Max)1,519.9002
9Gemini 3.8 Flash (High)1,519.4363
10Claude Opus 4.61,518.6692
11Qwen3.8-Max1,515.3131
12Claude Opus 5.5 (High)1,515.1587
13Claude Opus 5 (Max)1,514.7284
14Claude Opus 4.71,513.7216
15Gemini 3.7 Flash (High)1,512.8953
16Muse Spark 1.3 (Max)1,511.7716
17Muse Spark 1.2 (xHigh)1,510.1007
18Gemini 3.1 Pro Preview1,509.9619
19Claude Opus 5 (High)1,509.8956
20GLM-5.3 (Max)1,507.2569
21Claude Opus 4.8 (High)1,504.0575
22Muse Spark 1.11,501.3094
23Claude Opus 4.81,500.2422
24Claude Sonnet 4.61,499.5754
25Claude Sonnet 5.5 (xHigh)1,498.7925
26MiMo-V2.6-Pro1,498.7901
27GPT-5.5 (High)1,497.9711
28Gemini 3 Pro1,497.4934
29GLM-5.2 (Max)1,497.3676
30GPT-5.51,496.4947
31Gemini 3.5 Flash (Medium)1,496.1706
32Muse Spark1,495.4869
33Gemini 3.6 Flash (High)1,493.5919
34GPT-5.5 Instant1,491.9749
35Claude Opus 4.5 (High, 32K, 20251101)1,491.4778
36GPT-6 Astra (Max)1,490.7008
37GPT-6.1 Sol (Max)1,490.6421
38DeepSeek-V4.1-Flash-20260910 (Max)1,490.4787
39GLM-5.3-Flash1,490.2522
40Qwen3.5-Max-Preview1,490.079
41Gemini 3.5 Flash (High)1,490.0092
42MiMo-V2.5-Pro1,489.9926
43GPT-5.6 Sol (xHigh)1,489.911
44Grok 4.20 Beta (Reasoning, 0309)1,489.2671
45Claude Sonnet 5 (High)1,488.1131
46GPT-5.4 (High)1,487.7862
47DeepSeek-V4-Pro (High, 20260813)1,487.7389
48GLM-5.11,487.5271
49Claude Opus 4.5 (20251101)1,487.1886
50Gemini 3 Flash1,486.7469