Model Leaderboard

2171 models in the fused corpus; 1656 have enough benchmark coverage for an Aggregate Rating, human reference baselines excluded. Top 100 shown below.

RankModelProviderRatingBenchmarks
1Claude Mythos 5.1Anthropic2019 ± 2923
2Claude Mythos 5Anthropic2009 ± 13105
3Claude Opus 5.5Anthropic1986 ± 5713
4Claude Mythos PreviewAnthropic1984 ± 1297
5GPT-6 Astra ProOpenAI1984 ± 2120
6GPT-6 AstraOpenAI1983 ± 41009
7GPT-6.1 SolOpenAI1969 ± 6613
8Gemini 4 ArgonGoogle1967 ± 7429
9Claude Fable 5.1Anthropic1958 ± 5631
10Muse Wonder FullMeta1939 ± 20133
11Claude Sonnet 5.5Anthropic1934 ± 7482
12Claude Opus 5Anthropic1931 ± 31298
13GPT-6.1 Pro SolOpenAI1924 ± 2918
14GPT-6 SolOpenAI1910 ± 6577
15Muse WonderMeta1908 ± 9242
16Claude Fable 5Anthropic1907 ± 4914
17GPT-5.5 ProOpenAI1899 ± 1376
18GPT-5.6 Pro SolOpenAI1890 ± 1462
19GPT-5.6 SolOpenAI1889 ± 31958
20DeepSeek V4.1 Flash (1017)DeepSeek1874 ± 13230
21Gemini 3.8 FlashGoogle1873 ± 5865
22GPT-5.5 (2026-04-23)OpenAI1873 ± 14159
23Gemini 3.7 FlashGoogle1872 ± 5679
24Muse Spark 1.3Meta1866 ± 7423
25Grok 4.7xAI1865 ± 7439
26Grok 4.6xAI1864 ± 5593
27Gemini 3 Deep ThinkGoogle1864 ± 2221
28Qwen3.8 Omni FlashAlibaba1853 ± 17100
29Qwen 3.8 Max (0902)Alibaba1852 ± 11160
30Kimi K3Moonshot1851 ± 31284
31GPT-5.5OpenAI1850 ± 22986
32GPT-5.4 ProOpenAI1848 ± 1380
33MiniMax-M3.1-FlashMiniMax1848 ± 2475
34Doubao-Seed-2.1-Pro-260628ByteDance1847 ± 3632
35Qwen 3.8 MaxAlibaba1847 ± 5623
36DeepSeek V4.1 FlashDeepSeek1841 ± 5743
37Claude Opus 4.8Anthropic1839 ± 31910
38Hy4 previewTencent1838 ± 9249
39GLM-5.3Zhipu1836 ± 5774
40Claude Haiku 5.5Anthropic1836 ± 8394
41GPT-5.6 TerraOpenAI1835 ± 41083
42MiMo-V2.6-ProXiaomi1834 ± 9284
43Ember-1Fireworks AI1833 ± 2526
44Step 5 PreviewStepFun1828 ± 12191
45GPT-5.3 InstantOpenAI1827 ± 3129
46GPT-5 Pro (2025-10-06)OpenAI1827 ± 2220
47GPT-5.6 Pro LunaOpenAI1825 ± 2615
48Grok 4.5xAI1824 ± 5704
49Claude Opus 4.7Anthropic1824 ± 32129
50Qwen 3.8 FlashAlibaba1824 ± 9162
51Gemini 3.6 FlashGoogle1823 ± 5777
52Muse Spark 1.1Meta1821 ± 8307
53DeepSeek V4 Pro (0813)DeepSeek1819 ± 7451
54Muse Spark OpenMeta1819 ± 14233
55Qwen3.8 2.4T A95BAlibaba1817 ± 11259
56GPT-6 LunaOpenAI1814 ± 6634
57Claude Sonnet 5Anthropic1812 ± 41017
58Gemini 3.1 Pro (Preview)Google1810 ± 24642
59GPT-5.4 (2026-03-05)OpenAI1809 ± 8462
60Gemini 3.5 FlashGoogle1809 ± 31328
61GPT-5.2 ProOpenAI1807 ± 11152
62Muse Spark 1.2Meta1807 ± 7410
63Claude Opus 4.6Anthropic1806 ± 22841
64Muse SparkMeta1802 ± 15147
65GLM-5.3 FlashZhipu1800 ± 5759
66Gemini 3 Pro (Preview)Google1799 ± 41926
67Seed 2.1 ProByteDance1799 ± 9196
68GPT-5.6 LunaOpenAI1797 ± 41225
69GPT-5.3 CodexOpenAI1797 ± 5393
70MiMo-V2.6-FlashXiaomi1797 ± 11235
71Ling 3.1 FlashInclusionAI1796 ± 2375
72Qwen3.8-Flash-NextAlibaba1794 ± 11451
73GLM-5.2Zhipu1789 ± 31253
74Claude Opus 4.5 (20251101)Anthropic1787 ± 9239
75Qwen 3.7 MaxAlibaba1786 ± 4726
76DeepSeek-V4-Flash-Vision-ExpDeepSeek1785 ± 15169
77Doubao-Seed-2.0-Pro-260215ByteDance1784 ± 11122
78DeepSeek V4 Flash (0731)DeepSeek1784 ± 7442
79GPT-5.4OpenAI1783 ± 23594
80Gemini 3 ProGoogle1782 ± 31639
81GPT-5 ProOpenAI1781 ± 1378
82Gemini 3 Flash (Preview)Google1779 ± 31757
83Claude Opus 4.5Anthropic1779 ± 31369
84Qwen 3.5 Omni PlusAlibaba1774 ± 9245
85Qwen 3.6 Max PreviewAlibaba1773 ± 8375
86Seed 2.1 TurboByteDance1773 ± 1481
87Apodex 1.1Apodex1772 ± 2866
88Doubao Seed 2.0 LiteByteDance1772 ± 2118
89Kimi K2.7 CodeMoonshot1772 ± 7344
90Claude Sonnet 4.6Anthropic1769 ± 23725
91Beam 1.0Reflection1768 ± 2294
92Qwen 3.7 PlusAlibaba1768 ± 5580
93Hy3Tencent1768 ± 7402
94Mistral Large 4Mistral1768 ± 10235
95GPT-5.2 CodexOpenAI1767 ± 9189
96Qwen 3.6 MaxAlibaba1766 ± 1454
97GLM-5.1Zhipu1763 ± 31261
98GPT-5OpenAI1763 ± 24536
99Gemini 3 FlashGoogle1763 ± 22161
100GPT-5.1 Codex MaxOpenAI1763 ± 1564

Interactive version: theaggregate.ai/models · How It Works · Data refreshed daily, snapshot 2026-10-20.