← Artificial Analysis leaderboards

Humanity's Last Exam

Artificial Analysis · 50 models

RankModelScore
1Claude Opus 5.5 (Max, Default Fallback)0.614
2Claude Fable 5.1 (Max, Default Fallback)0.591
3Claude Fable 5.1 (Xhigh, Default Fallback)0.587
4Claude Opus 5.5 (Xhigh, Default Fallback)0.575
5Gemini 4 Argon (High)0.571
6Claude Fable 5.1 (High, Default Fallback)0.559
7Claude Opus 5.5 (High, Default Fallback)0.556
8Claude Fable 5 (Max, Opus 4.8 Fallback)0.555
9Claude Sonnet 5.5 (Max, Default Fallback)0.55
10Claude Opus 5 (Max)0.549
11Claude Opus 5.5 (Medium, Default Fallback)0.547
11GPT-6 Astra (Max)0.547
13GPT-6 Astra (Xhigh)0.546
14Claude Opus 5 (Xhigh)0.544
15Claude Fable 5.1 (Medium, Default Fallback)0.538
16Muse Wonder Full (Max)0.536
17GPT-6 Astra (High)0.531
18GPT-6.1 Sol (Max)0.529
19Claude Opus 5 (High)0.528
20GPT-6 Astra (Medium)0.527
21GPT-6.1 Sol (Xhigh)0.526
22Kimi K3.1 Pro0.525
23GPT-6.1 Sol (High)0.514
24Claude Opus 5 (Medium)0.513
25Muse Wonder (Max)0.501
26Claude Sonnet 5.5 (Xhigh, Default Fallback)0.5
27GPT-6.1 Sol (Medium)0.499
28GPT-5.6 Sol (Max)0.495
29MiMo-V2.6-Pro0.494
30GPT-6 Astra (Low)0.492
31Claude Fable 5.1 (Low, Default Fallback)0.489
32Claude Opus 4.8 (Max)0.487
32Muse Spark 1.3 (Max)0.487
34Claude Opus 5.5 (Low, Default Fallback)0.483
35MiniMax-M3.1-Flash0.48
36GPT-6 Sol (Max)0.479
36Gemini 3.7 Flash (High)0.479
38Gemini 3.8 Flash (High)0.478
39Muse Spark 1.3 (Xhigh)0.475
40GPT-6.1 Sol (Low)0.474
41GPT-5.6 Sol (Xhigh)0.473
42Gemini 3.1 Pro Preview0.47
43Kimi K3 (Max)0.469
44Step 5 Preview0.465
45GPT-6 Sol (Xhigh)0.463
46GPT-6.1 Terra (Max)0.462
46Muse Spark 1.1 (Xhigh)0.462
48GPT-5.6 Sol (High)0.46
49Claude Sonnet 5.5 (High, Default Fallback)0.458
49GPT-5.5 (Xhigh)0.458