Skip to main content
Radar

Five or fewer confirmed AI changes, with original sources, on mornings when something changed.A free source-linked morning brief.

See the free Radar Brief
BenchLM

Artificial Analysis Omniscience Index (AA-Omniscience Index)

A display-only Artificial Analysis factual knowledge index.

Data verified 23 confirmed releases in the last 30 daysSee the free Radar Brief

Benchmark score on AA-Omniscience Index — September 4, 2026

We mirror the published score view for AA-Omniscience Index. Claude Fable 5.1 leads the public snapshot at 43.5%, followed by GPT-6 Astra (43.4%) and Claude Fable 5 (43.3%). We do not use these results to rank models overall.

1Proprietary

Claude Fable 5.1

Anthropic

43.5%
Overall 82.95Context 1M
2Proprietary

GPT-6 Astra

OpenAI

43.4%
Overall 81.05Context 1.05M
3Proprietary

Claude Fable 5

Anthropic

43.3%
Overall 80.9Context 1M+

178 modelsKnowledgeCurrentDisplay onlyUpdated September 4, 2026

Benchmark score table (178 models)

Score
1
Claude Fable 5.1Anthropic · Closed
43.5%
2
GPT-6 AstraOpenAI · Closed
43.4%
3
Claude Fable 5Anthropic · Closed
43.3%
4
Claude Opus 5Anthropic · Closed
37.1%
5
Gemini 3.1 ProGoogle · Closed
31.9%
6
Grok 4.6xAI · Closed
30.5%
7
Gemini 3.8 FlashGoogle · Closed
29.6%
8
Claude Opus 4.8Anthropic · Closed
28.8%
9
Muse Spark 1.1Meta · Closed
28.1%
10
Claude Opus 4.7 (Adaptive)Anthropic · Closed
27.3%
11
Muse Spark 1.2Meta · Closed
27.2%
12
Gemini 3.7 FlashGoogle · Closed
26.5%
13
Grok 4.5xAI · Closed
25.3%
14
Muse Spark 1.3Meta · Closed
24.9%
15
Gemini 3.6 FlashGoogle · Closed
22.1%
16
GPT-5.6 SolOpenAI · Closed
22.0%
17
Gemini 3.5 FlashGoogle · Closed
21.2%
18
GPT-5.5OpenAI · Closed
20.5%
19
Kimi K3Moonshot AI · Closed
19.7%
20
Grok 4.3xAI · Closed
18.0%
21
Claude Sonnet 5Anthropic · Closed
16.5%
22
Gemini 3 ProGoogle · Closed
15.3%
23
Claude Opus 4.7Anthropic · Closed
14.8%
24
GLM-5.3Z.AI · Open weight
14.3%
25
Claude Opus 4.5 ThinkingAnthropic · Closed
14.0%
26
Claude Opus 4.6 (Adaptive)Anthropic · Closed
13.7%
27
Qwen3.7 MaxAlibaba · Closed
13.5%
28
GPT-5.3 CodexOpenAI · Closed
10.9%
29
GPT-5.3-Codex-SparkOpenAI · Closed
10.9%
30
Qwen 3.6 Max (preview)Alibaba · Closed
9.2%
31
GLM-5.3-FlashZ.AI · Open weight
7.5%
32
Muse SparkMeta · Closed
7.2%
33
GPT-5.4OpenAI · Closed
5.8%
34
GPT-5.1OpenAI · Closed
5.4%
35
Kimi K2.6Moonshot AI · Open weight
5.3%
36
Gemini 3.5 Flash-LiteGoogle · Closed
5.2%
37
MiMo-V2-ProXiaomi · Closed
4.6%
38
GLM-5.2Z.AI · Open weight
4.4%
39
Qwen3.8 Max PreviewAlibaba · Closed
3.4%
40
MiMo-V2.5-ProXiaomi · Closed
3.3%
41
Claude Opus 4.6Anthropic · Closed
2.4%
42
Grok 4xAI · Closed
2.1%
43
InklingThinking Machines Lab · Open weight
2.0%
44
MiniMax M3MiniMax · Open weight
1.4%
45
Qwen3.7 PlusAlibaba · Closed
1.1%
46
GLM-5.1Z.AI · Open weight
0.9%
47
Qwen3.6 PlusAlibaba · Closed
0.9%
48
DeepSeek V4 Pro 0813DeepSeek · Closed
0.8%
49
MiniMax M2.7MiniMax · Open weight
0.8%
50
GLM-5Z.AI · Open weight
0.3%
51
GPT-5.6 TerraOpenAI · Closed
0.1%
52
Nemotron 3 UltraNVIDIA · Open weight
-0.4%
53
GPT-5.2OpenAI · Closed
-0.9%
54
GPT-5.2-CodexOpenAI · Closed
-2.2%
55
Quasar 438BMultiverse Computing · Closed
-2.6%
56
Claude Sonnet 4.6Anthropic · Closed
-3.5%
57
Command A+Cohere · Open weight
-4.0%
58
Claude Opus 4.5Anthropic · Closed
-4.1%
59
Gemini 3 FlashGoogle · Closed
-4.3%
60
GPT-5.1-CodexOpenAI · Closed
-6.5%
61
GPT-5.1-Codex-MaxOpenAI · Closed
-6.5%
62
Kimi K2.5Moonshot AI · Open weight
-7.3%
63
Kimi K2.5 (Reasoning)Moonshot AI · Closed
-7.3%
64
GPT-5 (high)OpenAI · Closed
-8.7%
65
Inkling-SmallThinking Machines Lab · Open weight
-8.9%
66
Claude 4 SonnetAnthropic · Closed
-9.0%
67
Qwen3.8-Flash-NextAlibaba · Open weight
-9.7%
68
Qwen3.8-27BAlibaba · Open weight
-10.0%
69
Kimi K2.7 CodeMoonshot AI · Open weight
-10.2%
70
GPT-5.6 LunaOpenAI · Closed
-10.3%
71
GPT-4oOpenAI · Closed
-10.5%
72
DeepSeek V4 Pro (High)DeepSeek · Open weight
-10.6%
73
GPT-5 (medium)OpenAI · Closed
-10.9%
74
o1OpenAI · Closed
-11.0%
75
DeepSeek V4 Flash 0731DeepSeek · Closed
-14.3%
76
o3OpenAI · Closed
-15.5%
77
Gemini 2.5 ProGoogle · Closed
-16.3%
78
GLM-5-TurboZ.AI · Closed
-16.4%
79
Llama 3.1 405BMeta · Open weight
-17.1%
80
Granite 4.2 8BIBM · Open weight
-17.2%
81
GPT-5 miniOpenAI · Closed
-17.3%
82
Nemotron 3.5 Lightning 30B A3B NVFP4NVIDIA · Open weight
-17.7%
83
Ling 3.0 FlashInclusionAI · Open weight
-17.9%
84
Ling 3.0 Flash FP8InclusionAI · Open weight
-17.9%
85
Hy3Tencent · Open weight
-18.5%
86
Hy3 PreviewTencent · Open weight
-18.5%
87
GPT-5.4 miniOpenAI · Closed
-18.9%
88
GLM-5V-TurboZ.AI · Closed
-19.3%
89
Gemma 4 E4BGoogle · Open weight
-19.7%
90
Qwen3.6-27BAlibaba · Open weight
-20.0%
91
MiMo-V2-OmniXiaomi · Closed
-20.1%
92
Apodex 1.1Apodex · Closed
-21.9%
93
Apodex 1.1 MiniApodex · Open weight
-21.9%
94
Qwen3.6-35B-A3BAlibaba · Open weight
-22.2%
95
Gemma 4 E2BGoogle · Open weight
-23.6%
96
DeepSeek-R1DeepSeek · Open weight
-27.4%
97
Kimi K2Moonshot AI · Closed
-28.3%
98
GPT-5 nanoOpenAI · Closed
-28.7%
99
GPT-5.4 nanoOpenAI · Closed
-29.5%
100
LFM2.5-2.6BLiquidAI · Open weight
-29.5%
101
DeepSeek V3.1 (Reasoning)DeepSeek · Open weight
-29.6%
102
Grok 4.1 Fast (Reasoning)xAI · Closed
-29.9%
103
Grok 4 Fast (Reasoning)xAI · Closed
-29.9%
104
Mistral Small 4Mistral · Open weight
-30.4%
105
Mistral Small 4 (Reasoning)Mistral · Open weight
-30.4%
106
Qwen3.5 397BAlibaba · Open weight
-30.7%
107
Qwen3.5 397B (Reasoning)Alibaba · Open weight
-30.7%
108
Mistral Medium 3Mistral · Closed
-31.4%
109
GLM-4.6Z.AI · Open weight
-31.7%
110
Muse Glimmer 30BMeta · Open weight
-32.8%
111
LFM2.5-8B-A1BLiquidAI · Open weight
-33.3%
112
Mistral Large 2Mistral · Closed
-34.4%
113
GLM-4.7Z.AI · Open weight
-36.4%
114
Mistral Medium 3.5 128BMistral · Open weight
-36.8%
115
Grok Code Fast 1xAI · Closed
-37.1%
116
Step 3.7 FlashStepFun · Open weight
-37.3%
117
MiniMax M2.5MiniMax · Closed
-38.9%
118
Mistral Large 3Mistral · Closed
-39.6%
119
GPT-4.1OpenAI · Closed
-39.6%
120
Qwen3.5-122B-A10BAlibaba · Open weight
-41.5%
121
Nemotron 3 Super 100BNVIDIA · Open weight
-41.5%
122
Nemotron 3 Super 120B A12BNVIDIA · Open weight
-41.5%
123
DeepSeek V3DeepSeek · Open weight
-41.6%
124
Llama 4 MaverickMeta · Open weight
-41.8%
125
Gemini 2.5 FlashGoogle · Closed
-42.6%
126
DeepSeek V3.1DeepSeek · Open weight
-42.7%
127
Qwen3 MaxAlibaba · Closed
-43.5%
128
Qwen3.5-27BAlibaba · Open weight
-44.0%
129
Trinity-Large-ThinkingArcee AI · Open weight
-44.1%
130
Trinity-Large-PreviewArcee AI · Open weight
-44.1%
131
Step 3.5 FlashStepFun · Open weight
-44.2%
132
Nemotron Ultra 253BNVIDIA · Open weight
-44.9%
133
DeepSeek V3.2DeepSeek · Open weight
-46.9%
134
Nova ProAmazon · Closed
-47.7%
135
Gemma 4 31BGoogle · Open weight
-47.9%
136
Qwen3.5-35B-A3BAlibaba · Open weight
-48.1%
137
MiMo-V2-FlashXiaomi · Open weight
-48.4%
138
MiniMax M1 80kMiniMax · Closed
-48.5%
139
Claude 3 HaikuAnthropic · Closed
-48.6%
140
GPT-OSS 120BOpenAI · Open weight
-49.2%
141
Mercury 2.5 PreviewInception · Closed
-50.7%
142
Mercury 2Inception · Closed
-50.7%
143
Gemma 4 26B A4BGoogle · Open weight
-50.8%
144
Grok 4.1 FastxAI · Closed
-50.9%
145
Nemotron 3 Nano 30BNVIDIA · Open weight
-51.6%
146
Llama 4 ScoutMeta · Open weight
-52.1%
147
Gemma 4 12BGoogle · Open weight
-52.7%
148
GPT-4.1 miniOpenAI · Closed
-53.6%
149
Ultravox v0.6 Llama 3.3 70BFixie AI · Open weight
-54.2%
150
Phi-4Microsoft · Open weight
-55.7%
151
Nemotron 3 Nano Omni 30B A3BNVIDIA · Open weight
-57.4%
152
GPT-4.1 nanoOpenAI · Closed
-57.6%
153
K-ExaoneLG AI Research · Closed
-58.0%
154
LFM2-24B-A2BLiquidAI · Closed
-58.1%
155
Sarvam 105BSarvam · Open weight
-59.4%
156
GLM-4.5-AirZ.AI · Closed
-61.5%
157
Solar Pro 2Upstage · Closed
-62.0%
158
GLM-4.7-FlashZ.AI · Open weight
-62.6%
159
Exaone 4.0 32BLG AI Research · Open weight
-62.8%
160
GPT-OSS 20BOpenAI · Open weight
-63.0%
161
Ministral 3 3B (Reasoning)Mistral · Open weight
-64.0%
162
Ministral 3 3BMistral · Open weight
-64.0%
163
Ling 2.6 FlashInclusionAI · Open weight
-66.1%
164
Ministral 3 14B (Reasoning)Mistral · Open weight
-66.4%
165
Ministral 3 14BMistral · Open weight
-66.4%
166
Gemma 3 27BGoogle · Open weight
-67.2%
167
Ministral 3 8B (Reasoning)Mistral · Open weight
-68.9%
168
Ministral 3 8BMistral · Open weight
-68.9%
169
Granite-4.0-350MIBM · Open weight
-69.3%
170
Sarvam 30BSarvam · Open weight
-71.5%
171
Celeris-1Celeris · Closed
-71.6%
172
LFM2.5-1.2B-InstructLiquidAI · Closed
-72.1%
173
Granite-4.0-H-1BIBM · Open weight
-72.3%
174
LFM2.5-1.2B-ThinkingLiquidAI · Closed
-79.9%
175
Granite-4.0-H-350MIBM · Open weight
-80.9%
176
Granite-4.0-1BIBM · Open weight
-81.6%
177
Exaone 4.0 1.2BLG AI Research · Open weight
-82.1%
178
LFM2.5-VL-1.6B-ExtractLiquidAI · Open weight
-84.4%

The published AA-Omniscience Index snapshot places Claude Fable 5.1 first at 43.5%. The third row is 0.2 points behind. The broader top-10 range is 16.2 points, so the table still separates the published systems.

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

About AA-Omniscience Index

Year

2026

Tasks

Knowledge questions

Format

Index score

Difficulty

Broad factual knowledge

BenchLM stores the AA-Omniscience index as a display-only factuality signal alongside the accuracy and hallucination-rate rows.

AA-Omniscience: Knowledge and Hallucination Benchmark

BenchLM freshness & provenance

Version

AA-Omniscience Index 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-Omniscience Index measure?

A display-only Artificial Analysis factual knowledge index.

Which model scores highest on AA-Omniscience Index?

Claude Fable 5.1 by Anthropic currently leads with a score of 43.5% on AA-Omniscience Index.

How many models are evaluated on AA-Omniscience Index?

178 AI models have been evaluated on AA-Omniscience Index on BenchLM.

Compare Top Models on AA-Omniscience Index

Claude Fable 5.1 vs GPT-6 AstraGPT-6 Astra vs Claude Fable 5Claude Fable 5 vs Claude Opus 5Claude Opus 5 vs Gemini 3.1 Pro
Last updated: September 4, 2026 · BenchLM version AA-Omniscience Index 2026

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 2,000+ readers.

One email each week. Unsubscribe anytime.