← Arena leaderboards

Text · Factuality · Hard Prompts

Arena · 50 models

RankModelScore
1Gemini 4 Carbon (High)1,544.671
2Claude Opus 5.5 (High)1,537.2762
3Gemini 4 Argon (High)1,526.5921
4Claude Fable 5.1 (Max)1,524.3114
5Claude Opus 4.6 (High)1,522.2765
6Claude Opus 4.61,519.637
7Muse Wonder (Max)1,516.3637
8Claude Opus 5 (Max)1,515.2269
9GPT-5.5 (High)1,513.1526
10MiMo-V2.6-Pro1,512.5441
11GPT-5.51,510.8184
12GPT-5.4 (High)1,509.7355
13GPT-5.6 Sol (xHigh)1,509.1539
14Claude Fable 5 (High)1,508.1519
15Claude Opus 5 (High)1,506.0637
16Claude Sonnet 5.5 (xHigh)1,505.053
17Muse Spark 1.3 (Max)1,502.1106
18GPT-5.41,501.5466
19Claude Opus 4.7 (High)1,501.4462
20Muse Spark 1.2 (xHigh)1,500.7458
21Claude Opus 4.71,500.6483
22Qwen3.7-Max-Preview1,499.2231
23Claude Opus 4.8 (High)1,499.2124
24Kimi K3 (Max)1,499.1382
25Qwen3.5-Max-Preview1,499.0032
26GPT-6.1 Sol (Max)1,497.6975
27Gemini 3 Pro1,497.4149
28GPT-5.6 Terra (xHigh)1,497.3417
29GPT-6 Luna (Max)1,497.1429
30Gemini 3.7 Flash (High)1,496.938
31Gemini 3.5 Flash (High)1,496.4271
32DeepSeek-V4.1-Flash-20260910 (Max)1,495.9222
33GPT-6 Astra (Max)1,494.9112
34Qwen3.8-Max1,494.3828
35Gemini 3.8 Flash (High)1,493.3316
36MiMo-V2.5-Pro1,493.3169
37Claude Sonnet 4.61,492.7681
38Claude Opus 4.81,492.5423
39Claude Opus 4.5 (High, 32K, 20251101)1,492.1411
40Claude Opus 4.5 (20251101)1,491.7184
41Gemini 3.6 Flash (High)1,491.6479
42GLM-5.3-Flash1,491.4689
43Grok 4.51,491.0773
44GLM-5.3 (Max)1,490.7058
45Gemini 3.1 Pro Preview1,490.4085
46GPT-5.6 Luna (xHigh)1,488.6868
47Gemini 3.5 Flash (Medium)1,487.3713
48GLM-5.2 (Max)1,485.1437
49Muse Spark 1.11,483.8982
50GPT-5.1 (High)1,483.7034