← Arena leaderboards

Text · Factuality · Coding

Arena · 50 models

RankModelScore
1Gemini 4 Carbon (High)1,569.6789
2Claude Opus 5.5 (High)1,566.0705
3Claude Opus 4.6 (High)1,550.1901
4MiMo-V2.6-Pro1,547.4577
5Claude Fable 5.1 (Max)1,547.1971
6Claude Opus 4.61,547.024
7Muse Wonder (Max)1,546.0361
8Claude Opus 5 (Max)1,545.8079
9Gemini 4 Argon (High)1,545.1329
10Claude Opus 5 (High)1,537.1051
11Claude Fable 5 (High)1,536.2794
12Claude Sonnet 5.5 (xHigh)1,536.1572
13Claude Opus 4.7 (High)1,534.2859
14GPT-5.4 (High)1,533.9834
15Claude Opus 4.5 (High, 32K, 20251101)1,533.1758
16Muse Spark 1.3 (Max)1,532.8928
17GPT-5.5 (High)1,532.5488
18Muse Spark 1.2 (xHigh)1,532.4502
19GPT-5.6 Sol (xHigh)1,532.3065
20Claude Opus 4.71,532.1848
21Kimi K3 (Max)1,531.9682
22Claude Opus 4.5 (20251101)1,531.3824
23GPT-6 Astra (Max)1,529.3718
24Claude Sonnet 4.61,528.9733
25GLM-5.3-Flash1,528.5202
26MiMo-V2.5-Pro1,528.4867
27GPT-6.1 Sol (Max)1,527.6675
28DeepSeek-V4.1-Flash-20260910 (Max)1,526.256
29GPT-5.51,525.1462
30Gemini 3.7 Flash (High)1,524.3464
31GPT-5.41,524.3375
32Qwen3.5-Max-Preview1,523.7056
33Qwen3.8-Max1,523.5025
34Claude Opus 4.8 (High)1,523.5004
35GPT-5.6 Terra (xHigh)1,522.0605
36Claude Sonnet 4.5 (20250929)1,521.5481
37MiMo-V2.6-Flash1,520.6119
38GLM-5.3 (Max)1,520.5725
39Gemini 3.8 Flash (High)1,520.4514
40Gemini 3.5 Flash (High)1,520.3346
41Beam 1.0 501B A23B1,520.1203
42Claude Opus 4.81,519.7668
43Gemini 3.6 Flash (High)1,519.6986
44Gemini 3 Pro1,519.0056
45Claude Sonnet 4.5 (High, 32K, 20250929)1,518.2645
46Grok 4.51,516.872
47GPT-6 Luna (Max)1,516.6316
48Muse Spark 1.11,516.5809
49Kimi K2.5 (Instant)1,515.7446
50Kimi K2.61,515.081