← Arena leaderboards

Text · Factuality · Instruction Following

Arena · 50 models

RankModelScore
1Claude Opus 5.5 (High)1,528.2186
2Gemini 4 Carbon (High)1,520.6073
3Gemini 4 Argon (High)1,515.1644
4Claude Fable 5.1 (Max)1,509.8582
5Claude Opus 4.6 (High)1,507.2964
6Claude Opus 5 (Max)1,501.1997
7Claude Opus 4.61,497.8053
8Muse Wonder (Max)1,495.3645
9Claude Opus 5 (High)1,495.2477
10Claude Fable 5 (High)1,494.5237
11GPT-5.6 Sol (xHigh)1,491.9949
12Claude Sonnet 5.5 (xHigh)1,491.3609
13GPT-5.5 (High)1,491.2244
14GPT-5.51,488.9536
15Claude Opus 4.7 (High)1,487.2944
16MiMo-V2.6-Pro1,484.1791
17GPT-5.4 (High)1,483.7657
18Claude Opus 4.5 (High, 32K, 20251101)1,483.4911
19GPT-6.1 Sol (Max)1,482.9323
20Claude Opus 4.71,482.4817
21Claude Opus 4.8 (High)1,482.3364
22Kimi K3 (Max)1,482.0951
23Muse Spark 1.3 (Max)1,479.8698
24DeepSeek-V4.1-Flash-20260910 (Max)1,479.7411
25Gemini 3.7 Flash (High)1,479.4841
26Qwen3.5-Max-Preview1,478.1552
27Claude Opus 4.5 (20251101)1,477.7207
28GLM-5.3 (Max)1,477.5219
29Gemini 3.8 Flash (High)1,476.0901
30Claude Sonnet 4.61,476.0419
31Qwen3.7-Max-Preview1,474.7309
32Qwen3.8-Max1,474.7269
33GLM-5.3-Flash1,474.3414
34MiMo-V2.5-Pro1,473.8694
35GPT-5.41,473.822
36GPT-5.6 Terra (xHigh)1,470.6523
37Gemini 3.5 Flash (High)1,470.4409
38Claude Opus 4.81,470.3908
39Muse Spark 1.2 (xHigh)1,470.3233
40GPT-6 Astra (Max)1,469.1964
41Gemini 3 Pro1,468.8914
42Gemini 3.6 Flash (High)1,468.684
43Claude Sonnet 4.5 (20250929)1,468.3581
44GLM-5.2 (Max)1,468.0582
45Gemini 3.1 Pro Preview1,466.6419
46Claude Sonnet 4.5 (High, 32K, 20250929)1,465.7241
47GPT-6 Luna (Max)1,464.8208
48Grok 4.51,464.1039
49Gemini 3.5 Flash (Medium)1,463.5333
50GPT-5.6 Luna (xHigh)1,462.5508