← Arena leaderboards

Text · Factuality · Hard Prompts (English)

Arena · 50 models

RankModelScore
1Claude Opus 5.5 (High)1,549.1006
2Gemini 4 Carbon (High)1,544.5752
3Gemini 4 Argon (High)1,535.9274
4Claude Opus 4.6 (High)1,533.4883
5Claude Fable 5.1 (Max)1,532.0243
6Claude Opus 4.61,530.0235
7Muse Wonder (Max)1,529.1779
8Claude Opus 5 (Max)1,525.9131
9MiMo-V2.6-Pro1,522.6883
10Claude Sonnet 5.5 (xHigh)1,521.3585
11Claude Fable 5 (High)1,520.1939
12Claude Opus 5 (High)1,518.3737
13Qwen3.8-Max1,515.3144
14GPT-5.51,514.973
15GPT-5.5 (High)1,513.121
16Muse Spark 1.3 (Max)1,513.1056
17Kimi K3 (Max)1,512.9322
18GPT-5.4 (High)1,511.6147
19Muse Spark 1.2 (xHigh)1,511.4223
20Claude Opus 4.7 (High)1,511.2544
21GPT-6.1 Sol (Max)1,510.115
22Claude Opus 4.8 (High)1,510.0125
23MiMo-V2.5-Pro1,509.315
24Claude Opus 4.71,509.1007
25Qwen3.5-Max-Preview1,508.9363
26GPT-5.6 Sol (xHigh)1,508.3273
27Claude Sonnet 4.61,507.5317
28DeepSeek-V4.1-Flash-20260910 (Max)1,506.4824
29GLM-5.3 (Max)1,506.0607
30GPT-5.41,503.4066
31MiMo-V2.6-Flash1,503.2102
32Claude Opus 4.5 (20251101)1,502.7341
33Gemini 3.8 Flash (High)1,501.83
34Gemini 3.7 Flash (High)1,501.7137
35GPT-5.6 Terra (xHigh)1,500.8676
36Gemini 3.5 Flash (High)1,500.7595
37GPT-6 Astra (Max)1,500.5942
38Claude Opus 4.81,500.4359
39Gemini 3.6 Flash (High)1,500.3797
40GLM-5.3-Flash1,500.239
41Claude Opus 4.5 (High, 32K, 20251101)1,499.9094
42Gemini 3 Pro1,498.7878
43GPT-5.6 Luna (xHigh)1,498.1779
44GLM-5.11,497.9631
45Claude Sonnet 4.5 (High, 32K, 20250929)1,497.8884
46Grok 4.51,497.1012
47GLM-5.2 (Max)1,496.2514
48Gemini 3.1 Pro Preview1,495.7637
49Claude Sonnet 4.5 (20250929)1,495.4375
50MiMo-V2-Pro1,494.4353