← Arena leaderboards

Text · Factuality · Legal & Government

Arena · 50 models

RankModelScore
1Gemini 4 Argon (High)1,530.0557
2Claude Opus 5 (High)1,523.9399
3Claude Opus 5 (Max)1,519.9977
4GPT-5.5 (High)1,518.5705
5Claude Opus 4.6 (High)1,516.8445
6Claude Opus 4.61,512.8629
7Claude Fable 5.1 (Max)1,511.5954
8GPT-5.4 (High)1,511.1984
9Claude Opus 5.5 (High)1,509.3338
10Claude Fable 5 (High)1,509.0483
11Claude Opus 4.7 (High)1,508.7411
12Kimi K3 (Max)1,505.0443
13Gemini 3 Pro1,504.988
14GPT-5.6 Sol (xHigh)1,504.5873
15GPT-5.51,502.3784
16Gemini 3.5 Flash (High)1,501.0071
17Gemini 3.7 Flash (High)1,498.6666
18GPT-5.41,498.468
19Qwen3.5-Max-Preview1,498.2815
20Claude Opus 4.71,497.5302
21GPT-6 Astra (Max)1,497.1345
22GPT-5.6 Terra (xHigh)1,496.3497
23Gemini 3.1 Pro Preview1,495.2859
24Claude Opus 4.8 (High)1,495.0987
25Muse Spark 1.3 (Max)1,494.4455
26Claude Opus 4.81,493.0672
27Qwen3.8-Max1,491.5144
28Gemini 3.5 Flash (Medium)1,491.1327
29Gemini 3.8 Flash (High)1,490.3079
30GLM-5.2 (Max)1,489.574
31Gemini 2.5 Pro1,486.6485
32Grok 4.51,485.5546
33GPT-5.6 Luna (xHigh)1,485.5166
34Gemini 3.6 Flash (High)1,483.6971
35Claude Opus 4.5 (20251101)1,482.6383
36Muse Spark 1.11,482.0148
37MiMo-V2.5-Pro1,481.5536
38Claude Sonnet 4.61,481.3128
39GPT-5.1 (High)1,480.5669
40DeepSeek-V4-Pro1,478.1315
41GLM-5.3 (Max)1,478.0995
42GPT-5.11,477.559
43Claude Opus 4.5 (High, 32K, 20251101)1,477.3651
44GLM-5.11,476.8769
45DeepSeek-V4-Pro (High, 20260813)1,475.9281
46Claude Sonnet 5 (High)1,475.653
47Gemini 3 Flash1,474.8709
48GLM-5.3-Flash1,472.4224
49Kimi K2.61,471.9581
50DeepSeek-V4-Pro Preview (High)1,470.709