← Arena leaderboards

Text · Factuality · Non-English

Arena · 50 models

RankModelScore
1Gemini 4 Carbon (High)1,512.695
2Claude Opus 5.5 (High)1,495.7646
3Gemini 4 Argon (High)1,492.0883
4Claude Fable 5.1 (Max)1,490.7309
5Muse Wonder (Max)1,482.8119
6Claude Opus 5 (Max)1,479.8872
7GPT-5.5 (High)1,478.994
8Claude Opus 4.6 (High)1,478.4796
9Qwen3.7-Max-Preview1,475.7589
10GPT-5.4 (High)1,475.4788
11Claude Opus 4.61,474.3647
12Gemini 3 Pro1,473.8308
13GPT-5.51,471.5554
14Claude Fable 5 (High)1,471.0472
15Gemini 3.7 Flash (High)1,469.0369
16Claude Opus 5 (High)1,468.5479
17Gemini 3.5 Flash (High)1,468.3821
18GPT-5.6 Sol (xHigh)1,467.4777
19Claude Opus 4.7 (High)1,467.1793
20Muse Spark 1.3 (Max)1,467.152
21Gemini 3.8 Flash (High)1,466.6893
22Qwen3.5-Max-Preview1,466.4606
23Claude Opus 4.71,465.8414
24GPT-5.41,465.1384
25MiMo-V2.6-Pro1,464.7653
26Muse Spark 1.2 (xHigh)1,464.7031
27Gemini 3.1 Pro Preview1,463.5136
28Gemini 3.6 Flash (High)1,461.2009
29Kimi K3 (Max)1,458.644
30Gemini 3.5 Flash (Medium)1,458.5514
31GPT-6 Luna (Max)1,458.4849
32Gemini 3 Flash1,458.4458
33Claude Sonnet 5.5 (xHigh)1,458.2167
34Claude Opus 4.8 (High)1,457.9038
35GPT-5.6 Terra (xHigh)1,457.2008
36Qwen3.8-Max1,456.9462
37GPT-6 Astra (Max)1,456.5433
38Muse Spark1,455.8737
39Grok 4.51,454.473
40DeepSeek-V4.1-Flash-20260910 (Max)1,454.119
41Claude Opus 4.81,453.4715
42GPT-6.1 Sol (Max)1,453.0142
43Gemini 2.5 Pro1,452.7012
44GLM-5.3-Flash1,452.677
45GLM-5.3 (Max)1,452.2743
46Muse Spark 1.11,451.5912
47GLM-5.2 (Max)1,449.7145
48MiMo-V2.5-Pro1,446.9675
49GPT-5.6 Luna (xHigh)1,445.1992
50Claude Opus 4.5 (20251101)1,444.609