← Artificial Analysis leaderboards

GPQA Diamond

Artificial Analysis · 50 models

RankModelScore
1GPT-6 Astra (Xhigh)0.963
2GPT-6 Astra (Max)0.961
3Gemini 3.8 Flash (High)0.953
4GPT-6 Astra (High)0.949
4Grok 4.6 (High)0.949
6Gemini 3.7 Flash (High)0.945
7GPT-5.6 Sol (Max)0.941
7Gemini 3.1 Pro Preview0.941
7Muse Spark 1.3 (Xhigh)0.941
10GPT-6 Astra (Medium)0.939
11Claude Fable 5.1 (Max, Default Fallback)0.937
11Claude Opus 5 (High)0.937
11Claude Opus 5 (Xhigh)0.937
14GPT-5.5 (Xhigh)0.935
14Gemini 3.8 Flash (Medium)0.935
14Grok 4.6 (Medium)0.935
14Grok 4.6 (Xhigh)0.935
14Kimi K3 (Max)0.935
14Muse Spark 1.3 (Max)0.935
14Qwen3.8 2.4T A95B0.935
21Claude Fable 5.1 (Xhigh, Default Fallback)0.934
22Claude Opus 5 (Max)0.932
22GPT-5.5 (High)0.932
24GPT-5.6 Sol (Xhigh)0.931
24GPT-6 Astra (Low)0.931
24Grok 4.5 (High)0.931
27MiniMax-M30.929
28DeepSeek V4 Pro 0813 (Max)0.928
28GPT-5.6 Sol (High)0.928
28Gemini 3.6 Flash (High)0.928
28Qwen3.8 Max (0902)0.928
32Qwen3.8 Max0.927
33Claude Fable 5 (Max, Opus 4.8 Fallback)0.926
33GPT-5.5 (Medium)0.926
33GPT-5.6 Sol (Medium)0.926
36GPT-5.6 Terra (Max)0.925
37Qwen3.7 Max0.923
37Qwen3.8-Flash-Next0.923
39Gemini 3.5 Flash (High)0.922
40Gemini 3.5 Flash (Medium)0.921
40Gemini 3.7 Flash (Medium)0.921
42Claude Opus 4.8 (Max)0.92
42GPT-5.4 (Xhigh)0.92
42Gemini 3.8 Flash (Low)0.92
45Claude Opus 5 (Medium)0.919
46GLM-5.3 (Max)0.917
47GPT-5.3 Codex (Xhigh)0.915
48Claude Opus 4.7 (Max)0.914
49DeepSeek V4 Flash Vision (Max)0.913
50GLM-5.3-Flash0.912