← Arena leaderboards

Text · Factuality · Writing & Literature & Language

Arena · 50 models

RankModelScore
1Gemini 4 Carbon (High)1,521.5951
2Claude Opus 5.5 (High)1,513.2336
3Gemini 4 Argon (High)1,506.7576
4Claude Fable 5.1 (Max)1,500.8484
5Claude Fable 5 (High)1,489.1609
6Claude Opus 4.6 (High)1,488.751
7Muse Wonder (Max)1,485.2775
8Claude Opus 5 (Max)1,484.255
9Claude Opus 4.61,480.82
10Claude Opus 4.7 (High)1,480.1627
11Gemini 3.5 Flash (High)1,476.6973
12Gemini 3 Pro1,476.2114
13Gemini 3.7 Flash (High)1,475.7363
14Claude Opus 5 (High)1,475.0923
15Gemini 3.8 Flash (High)1,474.3796
16Claude Opus 4.71,471.9532
17Claude Sonnet 5.5 (xHigh)1,471.4881
18Qwen3.8-Max1,470.6459
19Gemini 3.1 Pro Preview1,470.5966
20GPT-5.5 (High)1,470.5276
21GPT-5.51,469.203
22GPT-5.6 Sol (xHigh)1,469.1291
23Qwen3.5-Max-Preview1,468.0825
24Kimi K3 (Max)1,466.7486
25Muse Spark 1.3 (Max)1,466.2857
26Claude Opus 4.8 (High)1,466.0372
27GPT-5.4 (High)1,464.1462
28Gemini 3.6 Flash (High)1,463.2419
29Gemini 3.5 Flash (Medium)1,463.2073
30Gemini 3 Flash1,462.7607
31GLM-5.3 (Max)1,462.6882
32Muse Spark1,462.3988
33GPT-5.41,459.4045
34GLM-5.2 (Max)1,458.9092
35MiMo-V2.6-Pro1,457.302
36Claude Opus 4.81,456.0327
37Grok 4.51,455.1043
38Claude Opus 4.5 (20251101)1,454.4876
39GPT-6.1 Sol (Max)1,454.284
40GPT-6 Astra (Max)1,453.9441
41DeepSeek-V4.1-Flash-20260910 (Max)1,453.6795
42Muse Spark 1.2 (xHigh)1,453.5068
43GLM-5.11,452.3079
44GLM-5.3-Flash1,452.2962
45MiMo-V2.5-Pro1,452.1435
46Claude Opus 4.5 (High, 32K, 20251101)1,451.7919
47Claude Sonnet 4.5 (20250929)1,450.7942
48Gemini 2.5 Pro1,450.0831
49Claude Sonnet 4.61,449.5897
50DeepSeek-V4-Pro1,449.0985