← Arena leaderboards

Text · Factuality · Expert

Arena · 50 models

RankModelScore
1Claude Opus 5.5 (High)1,548.8165
2Claude Opus 5 (Max)1,542.646
3Claude Opus 4.6 (High)1,542.1202
4Gemini 4 Argon (High)1,537.7512
5Claude Fable 5 (High)1,535.4016
6Claude Sonnet 5.5 (xHigh)1,535.2566
7Claude Opus 5 (High)1,534.9429
8Claude Opus 4.61,534.6111
9GPT-5.5 (High)1,531.8867
10GPT-5.6 Sol (xHigh)1,531.682
11Claude Fable 5.1 (Max)1,529.6839
12GPT-5.4 (High)1,529.5103
13GPT-6.1 Sol (Max)1,528.1156
14GPT-5.51,523.7257
15DeepSeek-V4.1-Flash-20260910 (Max)1,522.5632
16Kimi K3 (Max)1,522.3261
17Claude Opus 4.71,519.4622
18Claude Opus 4.7 (High)1,519.2055
19GPT-5.6 Terra (xHigh)1,516.9225
20GLM-5.3 (Max)1,516.3845
21GLM-5.3-Flash1,515.4527
22Claude Opus 4.8 (High)1,515.4325
23Gemini 3.8 Flash (High)1,514.2493
24Gemini 3.7 Flash (High)1,512.9076
25GPT-6 Astra (Max)1,512.2169
26Muse Spark 1.3 (Max)1,510.186
27GPT-5.41,509.6473
28MiMo-V2.5-Pro1,509.6327
29MiMo-V2.6-Flash1,509.2116
30Claude Sonnet 4.61,508.5639
31GPT-5.6 Luna (xHigh)1,506.2385
32Claude Opus 4.5 (20251101)1,506.0169
33Qwen3.5-Max-Preview1,505.5272
34Gemini 3.5 Flash (High)1,504.1971
35Claude Opus 4.5 (High, 32K, 20251101)1,503.1784
36ERNIE 5.11,503.1687
37Claude Opus 4.81,502.5729
38Qwen3.8-Max1,501.9246
39Claude Sonnet 5 (High)1,500.8368
40GPT-6 Luna (Max)1,500.181
41Step 5 Preview1,499.8473
42Kimi K2.61,498.9527
43Gemini 3.6 Flash (High)1,497.8476
44Gemini 3.5 Flash (Medium)1,497.4098
45GLM-5.2 (Max)1,495.9453
46Gemini 3 Pro1,495.9037
47Muse Spark 1.2 (xHigh)1,495.5716
48Grok 4.51,494.5847
49Gemini 3.1 Pro Preview1,493.7702
50Claude Sonnet 4.5 (High, 32K, 20250929)1,493.7589