Skip to main content
BenchLM
Advertise

Benchmark profile

Artificial Analysis GPQA Diamond (AA-GPQA Diamond)

A display-only Artificial Analysis GPQA Diamond score.

Data verified

Benchmark score on AA-GPQA Diamond — July 23, 2026

BenchLM mirrors the published score view for AA-GPQA Diamond. GPT-5.6 Sol leads the public snapshot at 94.1% , followed by Gemini 3.1 Pro (94.1%) and Kimi K3 (93.5%). BenchLM does not use these results to rank models overall.

1Closed

GPT-5.6 Sol

OpenAI

gpt-5-6-sol

94.1%
Overall 81.96Context 1M
2Closed

Gemini 3.1 Pro

Google

gemini-3-1-pro

94.1%
Overall 55.3Context 1M
3Closed

Kimi K3

Moonshot AI

kimi-3

93.5%
Overall 80.96Context 1.05M
162 modelsKnowledgeCurrentDisplay onlyUpdated July 23, 2026

Benchmark score table (162 models)

Score
1
GPT-5.6 SolOpenAI · Closed
94.1%
2
Gemini 3.1 ProGoogle · Closed
94.1%
3
Kimi K3Moonshot AI · Closed
93.5%
4
GPT-5.5OpenAI · Closed
93.5%
5
Grok 4.5xAI · Closed
93.1%
6
MiniMax M3MiniMax · Open weight
92.9%
7
Gemini 3.6 FlashGoogle · Closed
92.8%
8
Claude Fable 5Anthropic · Closed
92.6%
9
GPT-5.6 TerraOpenAI · Closed
92.5%
10
Qwen3.7 MaxAlibaba · Closed
92.3%
11
Gemini 3.5 FlashGoogle · Closed
92.2%
12
Claude Opus 4.8Anthropic · Closed
92.0%
13
GPT-5.4OpenAI · Closed
92.0%
14
GPT-5.3 CodexOpenAI · Closed
91.5%
15
GPT-5.3-Codex-SparkOpenAI · Closed
91.5%
16
Claude Opus 4.7 (Adaptive)Anthropic · Closed
91.4%
17
Claude Sonnet 5Anthropic · Closed
91.1%
18
GPT-5.6 LunaOpenAI · Closed
91.1%
19
Kimi K2.6Moonshot AI · Open weight
91.1%
20
Gemini 3 ProGoogle · Closed
90.8%
21
DeepSeek V4 Pro (High)DeepSeek · Open weight
90.5%
22
GPT-5.2OpenAI · Closed
90.3%
23
Grok 4.3xAI · Closed
90.1%
24
Qwen3.7 PlusAlibaba · Closed
90.0%
25
GPT-5.2-CodexOpenAI · Closed
89.9%
26
Muse Spark 1.1Meta · Closed
89.8%
27
Hy3 PreviewTencent · Open weight
89.7%
28
Hy3Tencent · Open weight
89.7%
29
Claude Opus 4.6 (Adaptive)Anthropic · Closed
89.6%
30
Kimi K2.7 CodeMoonshot AI · Open weight
89.6%
31
GLM-5.2Z.AI · Open weight
89.5%
32
DeepSeek V4 Flash (Max)DeepSeek · Open weight
89.4%
33
Qwen3.5 397BAlibaba · Open weight
89.3%
34
Qwen3.5 397B (Reasoning)Alibaba · Open weight
89.3%
35
DeepSeek V4 Pro (Max)DeepSeek · Open weight
88.8%
36
Qwen 3.6 Max (preview)Alibaba · Closed
88.8%
37
Claude Opus 4.7Anthropic · Closed
88.5%
38
Muse SparkMeta · Closed
88.4%
39
Qwen3.6 PlusAlibaba · Closed
88.2%
40
Kimi K2.5Moonshot AI · Open weight
87.9%
41
Kimi K2.5 (Reasoning)Moonshot AI · Closed
87.9%
42
Grok 4xAI · Closed
87.7%
43
GPT-5.4 miniOpenAI · Closed
87.5%
44
MiniMax M2.7MiniMax · Open weight
87.4%
45
GPT-5.1OpenAI · Closed
87.3%
46
InklingThinking Machines Lab · Open weight
87.2%
47
MiMo-V2-ProXiaomi · Closed
87.0%
48
GLM-5.1Z.AI · Open weight
86.8%
49
Nemotron 3 UltraNVIDIA · Open weight
86.7%
50
DeepSeek V4 Flash (High)DeepSeek · Open weight
86.7%
51
MiMo-V2.5-ProXiaomi · Closed
86.6%
52
Claude Opus 4.5 ThinkingAnthropic · Closed
86.6%
53
GPT-5.1-Codex-MaxOpenAI · Closed
86.0%
54
GPT-5.1-CodexOpenAI · Closed
86.0%
55
GLM-4.7Z.AI · Open weight
85.9%
56
Qwen3.5-27BAlibaba · Open weight
85.8%
57
Qwen3.5-122B-A10BAlibaba · Open weight
85.7%
58
Gemma 4 31BGoogle · Open weight
85.7%
59
GPT-5 (high)OpenAI · Closed
85.4%
60
Grok 4.1 Fast (Reasoning)xAI · Closed
85.3%
61
MiniMax M2.5MiniMax · Closed
84.8%
62
GLM-5-TurboZ.AI · Closed
84.7%
63
Grok 4 Fast (Reasoning)xAI · Closed
84.7%
64
Qwen3.5-35B-A3BAlibaba · Open weight
84.5%
65
o3-proOpenAI · Closed
84.5%
66
Gemini 2.5 ProGoogle · Closed
84.4%
67
Qwen3.6-27BAlibaba · Open weight
84.2%
68
GPT-5 (medium)OpenAI · Closed
84.2%
69
Qwen3.6-35B-A3BAlibaba · Open weight
84.1%
70
Claude Opus 4.6Anthropic · Closed
84.0%
71
Gemini 3.5 Flash-LiteGoogle · Closed
83.8%
72
MiMo-V2-OmniXiaomi · Closed
82.8%
73
GPT-5 miniOpenAI · Closed
82.8%
74
o3OpenAI · Closed
82.7%
75
Step 3.5 FlashStepFun · Open weight
82.6%
76
Gemini 3.1 Flash-LiteGoogle · Closed
82.2%
77
GLM-5Z.AI · Open weight
82.0%
78
GPT-5.4 nanoOpenAI · Closed
81.7%
79
DeepSeek-R1DeepSeek · Open weight
81.3%
80
Gemini 3 FlashGoogle · Closed
81.2%
81
Claude Opus 4.5Anthropic · Closed
81.0%
82
Step 3.7 FlashStepFun · Open weight
80.9%
83
GLM-5V-TurboZ.AI · Closed
80.9%
84
Claude 4.1 Opus ThinkingAnthropic · Closed
80.9%
85
Nemotron 3 Super 120B A12BNVIDIA · Open weight
80.0%
86
Claude Sonnet 4.6Anthropic · Closed
79.9%
87
Gemma 4 26B A4BGoogle · Open weight
79.2%
88
K-ExaoneLG AI Research · Closed
78.3%
89
GPT-OSS 120BOpenAI · Open weight
78.2%
90
DeepSeek V3.1 (Reasoning)DeepSeek · Open weight
77.9%
91
Mercury 2Inception · Closed
77.0%
92
Mistral Small 4Mistral · Open weight
76.9%
93
Mistral Small 4 (Reasoning)Mistral · Open weight
76.9%
94
Kimi K2Moonshot AI · Closed
76.6%
95
Qwen3 MaxAlibaba · Closed
76.4%
96
Command A+Cohere · Open weight
76.1%
97
Nemotron 3 Nano 30BNVIDIA · Open weight
75.7%
98
Gemma 4 12BGoogle · Open weight
75.3%
99
Trinity-Large-PreviewArcee AI · Open weight
75.2%
100
Trinity-Large-ThinkingArcee AI · Open weight
75.2%
101
DeepSeek V3.2DeepSeek · Open weight
75.1%
102
Mistral Medium 3.5 128BMistral · Open weight
74.8%
103
o3-miniOpenAI · Closed
74.8%
104
o1OpenAI · Closed
74.7%
105
Sarvam 105BSarvam · Open weight
73.8%
106
DeepSeek V3.1DeepSeek · Open weight
73.5%
107
GLM-4.5-AirZ.AI · Closed
73.3%
108
Nemotron Ultra 253BNVIDIA · Open weight
72.8%
109
Grok Code Fast 1xAI · Closed
72.7%
110
MiniMax M1 80kMiniMax · Closed
69.7%
111
GPT-OSS 20BOpenAI · Open weight
68.8%
112
Claude 4 SonnetAnthropic · Closed
68.3%
113
Gemini 2.5 FlashGoogle · Closed
68.3%
114
Mistral Large 3Mistral · Closed
68.0%
115
GPT-5 nanoOpenAI · Closed
67.6%
116
Llama 4 MaverickMeta · Open weight
67.1%
117
GPT-4.1OpenAI · Closed
66.6%
118
GPT-4.1 miniOpenAI · Closed
66.4%
119
MiMo-V2-FlashXiaomi · Open weight
65.6%
120
Grok 4.1 FastxAI · Closed
63.7%
121
Sarvam 30BSarvam · Open weight
63.3%
122
GLM-4.6Z.AI · Open weight
63.2%
123
Exaone 4.0 32BLG AI Research · Open weight
62.8%
124
DeepSeek R1 Distill Qwen 32BDeepSeek · Open weight
61.5%
125
Ling 2.6 FlashInclusionAI · Open weight
59.3%
126
Gemini 1.5 ProGoogle · Closed
58.9%
127
Llama 4 ScoutMeta · Open weight
58.7%
128
GLM-4.7-FlashZ.AI · Open weight
58.1%
129
Mistral Medium 3Mistral · Closed
57.8%
130
Gemma 4 E4BGoogle · Open weight
57.6%
131
Phi-4Microsoft · Open weight
57.5%
132
Ministral 3 14B (Reasoning)Mistral · Open weight
57.2%
133
Ministral 3 14BMistral · Open weight
57.2%
134
Solar Pro 2Upstage · Closed
56.1%
135
DeepSeek V3DeepSeek · Open weight
55.7%
136
GPT-4oOpenAI · Closed
54.3%
137
Llama 3.1 405BMeta · Open weight
51.5%
138
LFM2.5-8B-A1BLiquidAI · Open weight
51.3%
139
GPT-4.1 nanoOpenAI · Closed
51.2%
140
Nova ProAmazon · Closed
49.9%
141
Claude 3 OpusAnthropic · Closed
48.9%
142
Mistral Large 2Mistral · Closed
48.6%
143
LFM2-24B-A2BLiquidAI · Closed
47.4%
144
Ministral 3 8B (Reasoning)Mistral · Open weight
47.1%
145
Ministral 3 8BMistral · Open weight
47.1%
146
Nemotron 3 Nano Omni 30B A3BNVIDIA · Open weight
46.9%
147
Gemma 3 27BGoogle · Open weight
42.8%
148
GPT-4o miniOpenAI · Closed
42.6%
149
Exaone 4.0 1.2BLG AI Research · Open weight
42.4%
150
Qwen2.5 Coder 32B InstructAlibaba · Open weight
41.7%
151
Gemma 4 E2BGoogle · Open weight
37.5%
152
Claude 3 HaikuAnthropic · Closed
37.4%
153
Ministral 3 3B (Reasoning)Mistral · Open weight
35.8%
154
Ministral 3 3BMistral · Open weight
35.8%
155
LFM2.5-1.2B-ThinkingLiquidAI · Closed
33.9%
156
LFM2.5-1.2B-InstructLiquidAI · Closed
32.6%
157
LFM2.5-VL-1.6B-ExtractLiquidAI · Open weight
28.9%
158
Granite-4.0-1BIBM · Open weight
28.1%
159
Gemini 1.0 ProGoogle · Closed
27.7%
160
Granite-4.0-H-1BIBM · Open weight
26.3%
161
Granite-4.0-H-350MIBM · Open weight
25.7%
162
Granite-4.0-350MIBM · Open weight
20.3%

The published AA-GPQA Diamond snapshot places GPT-5.6 Sol first at 94.1%. The third row is 0.6 points behind. The broader top-10 range is 1.8 points, so many of the published results sit in a relatively narrow band.

162 models have been evaluated on AA-GPQA Diamond. The benchmark falls in the Knowledge category. This category carries a 12% weight in BenchLM.ai's overall scoring system. AA-GPQA Diamond is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About AA-GPQA Diamond

Year

2026

Tasks

Graduate-level science questions

Format

Accuracy

Difficulty

Graduate-level science reasoning

BenchLM stores the Artificial Analysis GPQA Diamond result separately from the weighted GPQA lane so AA refreshes remain display-only.

Artificial Analysis GPQA Diamond Benchmark Leaderboard

BenchLM freshness & provenance

Version

AA-GPQA Diamond 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

FAQ

What does AA-GPQA Diamond measure?

A display-only Artificial Analysis GPQA Diamond score.

Which model scores highest on AA-GPQA Diamond?

GPT-5.6 Sol by OpenAI currently leads with a score of 94.1% on AA-GPQA Diamond.

How many models are evaluated on AA-GPQA Diamond?

162 AI models have been evaluated on AA-GPQA Diamond on BenchLM.

Compare Top Models on AA-GPQA Diamond

GPT-5.6 Sol vs Gemini 3.1 ProGemini 3.1 Pro vs Kimi K3Kimi K3 vs GPT-5.5GPT-5.5 vs Grok 4.5
Last updated: July 23, 2026 · BenchLM version AA-GPQA Diamond 2026

Choose a model with this week’s evidence

Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.

One email each week. Unsubscribe anytime.