← Arena leaderboards

Text · Factuality · Creative Writing

Arena · 50 models

RankModelScore
1Claude Opus 5.5 (High)1,509.3903
2Gemini 4 Argon (High)1,504.9417
3Gemini 4 Carbon (High)1,500.8657
4Claude Fable 5.1 (Max)1,495.2088
5Claude Opus 4.6 (High)1,490.3844
6Muse Wonder (Max)1,480.42
7Gemini 3 Pro1,478.7568
8Claude Opus 5 (Max)1,478.3028
9Claude Fable 5 (High)1,477.7146
10Claude Opus 4.7 (High)1,472.7427
11Gemini 3.7 Flash (High)1,471.5853
12Gemini 3.8 Flash (High)1,469.3496
13Claude Opus 5 (High)1,468.7682
14Claude Opus 4.61,468.5358
15Claude Opus 4.71,467.5873
16Gemini 3.5 Flash (High)1,467.0642
17Qwen3.8-Max1,466.229
18Gemini 3.1 Pro Preview1,465.5152
19Gemini 3 Flash1,461.9443
20Claude Sonnet 5.5 (xHigh)1,460.9625
21Gemini 3.5 Flash (Medium)1,460.8852
22Muse Spark 1.3 (Max)1,459.9251
23Qwen3.5-Max-Preview1,459.9176
24MiMo-V2.6-Pro1,457.4902
25GPT-5.5 (High)1,456.5245
26Gemini 3.6 Flash (High)1,455.9754
27GLM-5.3 (Max)1,455.5956
28GPT-5.51,454.7828
29Claude Opus 4.8 (High)1,454.7813
30GPT-5.6 Sol (xHigh)1,453.7274
31GLM-5.2 (Max)1,451.5432
32Kimi K3 (Max)1,451.4193
33Claude Opus 4.81,448.5168
34Claude Opus 4.5 (20251101)1,448.4715
35GPT-5.4 (High)1,447.9948
36Gemini 2.5 Pro1,447.9944
37GLM-5.11,447.9567
38Grok 4.51,447.5013
39Claude Sonnet 4.5 (20250929)1,446.4688
40DeepSeek-V4-Pro1,445.8136
41Claude Opus 4.5 (High, 32K, 20251101)1,445.4979
42ERNIE 5.11,443.301
43GLM-51,442.8215
44Grok 4.20 Multi-Agent Beta (0309)1,442.4674
45Claude Sonnet 4.61,442.173
46Grok 4.20 Beta 11,441.892
47MiMo-V2.5-Pro1,441.5375
48DeepSeek-V4-Pro Preview (High)1,439.8107
49GPT-5.41,439.0706
50DeepSeek-V4.1-Flash-20260910 (Max)1,438.44