Artificial Analysis Omniscience Index (AA-Omniscience Index)
We show this table for reference; we do not rank on it.
Data verified
A display-only Artificial Analysis factual knowledge index.
Benchmark score on AA-Omniscience Index — November 7, 2026
We compile the AA-Omniscience Index rows from provider self-reports and secondary reports. Gemini 4 Carbon leads the table at 53.5%, followed by Claude Opus 5.5 (48.1%) and Claude Fable 5.1 (46.3%). We do not use these results to rank models overall.
1Closed
Gemini 4 Carbon
53.5%
Overall 88.69
2Closed
Claude Opus 5.5
Anthropic
48.1%
Overall 86.32Context 1M
3Closed
Claude Fable 5.1
Anthropic
46.3%
Overall 81.3Context 1M
167 modelsKnowledgeCurrentDisplay onlyUpdated November 7, 2026
Benchmark score table (167 models)
167 of 167 rows
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | Gemini 4 CarbonGoogle | 53.5% | Not reported | Closed |
| 2 | Claude Opus 5.5Anthropic | 48.1% | Not reported | Closed |
| 3 | Claude Fable 5.1Anthropic | 46.3% | Not reported | Closed |
| 4 | Gemini 4 ArgonGoogle | 46.1% | Not reported | Closed |
| 5 | Claude Fable 5Anthropic | 43.3% | Not reported | Closed |
| 6 | GPT-6 AstraOpenAI | 42.4% | Not reported | Closed |
| 7 | GPT-6.1 SolOpenAI | 42.0% | Not reported | Closed |
| 8 | Claude Opus 5Anthropic | 37.1% | Not reported | Closed |
| 9 | Grok 4.7xAI | 31.2% | Not reported | Closed |
| 10 | Muse Wonder FullMeta | 30.6% | Not reported | Closed |
| 11 | Claude Sonnet 5.5Anthropic | 30.5% | Not reported | Closed |
| 12 | Grok 4.6xAI | 30.5% | Not reported | Closed |
| 13 | Kimi K3.1 ProMoonshot AI | 29.9% | Not reported | Open |
| 14 | GPT-6.1 TerraOpenAI | 29.6% | Not reported | Closed |
| 15 | Claude Opus 4.8Anthropic | 28.8% | Not reported | Closed |
| 16 | Muse Spark 1.1Meta | 28.1% | Not reported | Closed |
| 17 | Claude Opus 4.7 (Adaptive)Anthropic | 27.3% | Not reported | Closed |
| 18 | Muse Spark 1.2Meta | 27.2% | Not reported | Closed |
| 19 | GPT-6 SolOpenAI | 27.1% | Not reported | Closed |
| 20 | Gemini 3.7 FlashGoogle | 26.5% | Not reported | Closed |
| 21 | Muse Spark OpenMeta | 25.8% | Not reported | Open |
| 22 | Grok 4.5xAI | 25.3% | Not reported | Closed |
| 23 | Muse WonderMeta | 25.1% | Not reported | Closed |
| 24 | Muse Spark 1.3Meta | 25.0% | Not reported | Closed |
| 25 | Gemini 3.8 FlashGoogle | 23.3% | Not reported | Closed |
| 26 | Gemini 3.6 FlashGoogle | 22.1% | Not reported | Closed |
| 27 | GPT-5.6 SolOpenAI | 22.0% | Not reported | Closed |
| 28 | Gemini 3.5 FlashGoogle | 21.2% | Not reported | Closed |
| 29 | GPT-5.5OpenAI | 20.5% | Not reported | Closed |
| 30 | Kimi K3Moonshot AI | 19.7% | Not reported | Open |
| 31 | Grok 4.3xAI | 18.0% | Not reported | Closed |
| 32 | Claude Sonnet 5Anthropic | 16.5% | Not reported | Closed |
| 33 | Step 5 PreviewStepFun | 16.4% | Not reported | Open |
| 34 | Inkling 2 SmallThinking Machines Lab | 16.1% | Not reported | Closed |
| 35 | Gemini 3 ProGoogle | 15.3% | Not reported | Closed |
| 36 | Claude Opus 4.7Anthropic | 14.8% | Not reported | Closed |
| 37 | Claude Opus 4.5 ThinkingAnthropic | 14.0% | Not reported | Closed |
| 38 | Claude Opus 4.6 (Adaptive)Anthropic | 13.7% | Not reported | Closed |
| 39 | Qwen3.7 MaxAlibaba | 13.5% | Not reported | Closed |
| 40 | Claude Haiku 5.5Anthropic | 12.3% | Not reported | Closed |
| 41 | Qwen3.8 Max PreviewAlibaba | 10.5% | Not reported | Closed |
| 42 | Qwen 3.6 Max (preview)Alibaba | 9.2% | Not reported | Closed |
| 43 | Muse SparkMeta | 7.2% | Not reported | Closed |
| 44 | Gemini 3.5 Flash-LiteGoogle | 6.9% | Not reported | Closed |
| 45 | GLM-5.3Z.AI | 6.0% | Not reported | Open |
| 46 | GPT-5.4OpenAI | 5.8% | Not reported | Closed |
| 47 | GPT-5.1OpenAI | 5.4% | Not reported | Closed |
| 48 | Kimi K2.6Moonshot AI | 5.3% | Not reported | Open |
| 49 | Nemotron 3.5 Super VL 120B A12B NVFP4NVIDIA | 5.3% | Not reported | Open |
| 50 | Mistral Large 4Mistral | 5.1% | Not reported | Open |
| 51 | MiMo-V2-ProXiaomi | 4.6% | Not reported | Closed |
| 52 | GLM-5.2Z.AI | 4.4% | Not reported | Open |
| 53 | Nemotron 3.5 Super VL 120B A12BNVIDIA | 4.0% | Not reported | Open |
| 54 | Nemotron 3 UltraNVIDIA | 3.4% | Not reported | Open |
| 55 | MiMo-V2.5-ProXiaomi | 3.3% | Not reported | Closed |
| 56 | Claude Opus 4.6Anthropic | 2.4% | Not reported | Closed |
| 57 | Beam 1.0 501B-A23BReflection AI | 2.3% | Not reported | Open |
| 58 | Grok 4xAI | 2.1% | Not reported | Closed |
| 59 | Solar Pro 4Upstage | 1.4% | Not reported | Closed |
| 60 | Qwen3.6 PlusAlibaba | 0.9% | Not reported | Closed |
| 61 | GLM-5.1Z.AI | 0.9% | Not reported | Open |
| 62 | DeepSeek V4 Pro 0813DeepSeek | 0.8% | Not reported | Open |
| 63 | MiniMax M2.7MiniMax | 0.8% | Not reported | Open |
| 64 | LFM3-1.7BLiquidAI | 0.5% | Not reported | Open |
| 65 | GLM-5Z.AI | 0.3% | Not reported | Open |
| 66 | GPT-5.6 TerraOpenAI | 0.1% | Not reported | Closed |
| 67 | Claude 3.7 Sonnet ThinkingAnthropic | -0.7% | Not reported | Closed |
| 68 | GPT-5.2OpenAI | -0.9% | Not reported | Closed |
| 69 | InklingThinking Machines Lab | -1.5% | Not reported | Open |
| 70 | GPT-5.2-CodexOpenAI | -2.2% | Not reported | Closed |
| 71 | Command A+Cohere | -2.7% | Not reported | Open |
| 72 | Claude Sonnet 4.6Anthropic | -3.5% | Not reported | Closed |
| 73 | Claude Opus 4.5Anthropic | -4.1% | Not reported | Closed |
| 74 | Gemini 3 FlashGoogle | -4.3% | Not reported | Closed |
| 75 | MiMo-V2.6-ProXiaomi | -4.5% | Not reported | Open |
| 76 | DeepSeek V4.1 Flash 1017DeepSeek | -5.3% | Not reported | Open |
| 77 | MiniMax M3MiniMax | -5.5% | Not reported | Open |
| 78 | MiniMax M3.1 FlashMiniMax | -6.4% | Not reported | Closed |
| 79 | Qwen3.8-27BAlibaba | -6.4% | Not reported | Open |
| 80 | GPT-5.1-CodexOpenAI | -6.5% | Not reported | Closed |
| 81 | GPT-5.1-Codex-MaxOpenAI | -6.5% | Not reported | Closed |
| 82 | Kimi K2.5 (Reasoning)Moonshot AI | -7.3% | Not reported | Closed |
| 83 | Kimi K2.5Moonshot AI | -7.3% | Not reported | Open |
| 84 | GPT-6 LunaOpenAI | -8.3% | Not reported | Closed |
| 85 | GPT-5 (high)OpenAI | -8.7% | Not reported | Closed |
| 86 | Claude 4 SonnetAnthropic | -9.0% | Not reported | Closed |
| 87 | Inkling-SmallThinking Machines Lab | -9.5% | Not reported | Open |
| 88 | Claude 3.7 SonnetAnthropic | -9.7% | Not reported | Closed |
| 89 | GPT-5.6 LunaOpenAI | -10.3% | Not reported | Closed |
| 90 | GPT-4oOpenAI | -10.5% | Not reported | Closed |
| 91 | DeepSeek V4 Pro (High)DeepSeek | -10.6% | Not reported | Open |
| 92 | MiMo-V2.6-FlashXiaomi | -10.8% | Not reported | Open |
| 93 | GPT-5 (medium)OpenAI | -10.9% | Not reported | Closed |
| 94 | o1OpenAI | -11.0% | Not reported | Closed |
| 95 | K-EXAONE 2.0LG AI Research | -11.5% | Not reported | Open |
| 96 | Qwen3.8-Flash-NextAlibaba | -12.2% | Not reported | Open |
| 97 | Granite 4.2 30BIBM | -12.4% | Not reported | Open |
| 98 | DeepSeek V4 Flash 0731DeepSeek | -14.3% | Not reported | Open |
| 99 | Nemotron 3.5 Lightning 30B A3B NVFP4NVIDIA | -15.7% | Not reported | Open |
| 100 | Ling 3.0 TinyInclusionAI | -15.9% | Not reported | Open |
| 101 | Gemini 2.5 ProGoogle | -16.3% | Not reported | Closed |
| 102 | GLM-5-TurboZ.AI | -16.4% | Not reported | Closed |
| 103 | Granite 4.2 8BIBM | -16.7% | Not reported | Open |
| 104 | GPT-5 miniOpenAI | -17.3% | Not reported | Closed |
| 105 | Ling 3.0 FlashInclusionAI | -17.9% | Not reported | Open |
| 106 | Granite 4.2 3BIBM | -18.3% | Not reported | Open |
| 107 | GPT-5.4 miniOpenAI | -18.9% | Not reported | Closed |
| 108 | GLM-5V-TurboZ.AI | -19.3% | Not reported | Closed |
| 109 | Qwen3.6-27BAlibaba | -20.0% | Not reported | Open |
| 110 | MiMo-V2-OmniXiaomi | -20.1% | Not reported | Closed |
| 111 | Claude 3.5 HaikuAnthropic | -22.5% | Not reported | Closed |
| 112 | LFM2.5-24B-A2BLiquidAI | -24.9% | Not reported | Open |
| 113 | Magistral MediumMistral | -26.9% | Not reported | Closed |
| 114 | DeepSeek-R1DeepSeek | -27.4% | Not reported | Open |
| 115 | Kimi K2Moonshot AI | -28.3% | Not reported | Closed |
| 116 | GPT-5 nanoOpenAI | -28.7% | Not reported | Closed |
| 117 | GPT-5.4 nanoOpenAI | -29.5% | Not reported | Closed |
| 118 | LFM2.5-2.6BLiquidAI | -29.5% | Not reported | Open |
| 119 | DeepSeek V3.1 (Reasoning)DeepSeek | -29.6% | Not reported | Open |
| 120 | Grok 4.1 Fast (Reasoning)xAI | -29.9% | Not reported | Closed |
| 121 | Grok 4 Fast (Reasoning)xAI | -29.9% | Not reported | Closed |
| 122 | Mistral Small 4Mistral | -31.0% | Not reported | Open |
| 123 | Mistral Small 4 (Reasoning)Mistral | -31.0% | Not reported | Open |
| 124 | Mistral Medium 3Mistral | -31.4% | Not reported | Closed |
| 125 | GLM-4.6Z.AI | -31.7% | Not reported | Open |
| 126 | Muse Glimmer 30BMeta | -32.2% | Not reported | Open |
| 127 | LFM2.5-8B-A1BLiquidAI | -33.3% | Not reported | Open |
| 128 | Mistral Large 2Mistral | -34.4% | Not reported | Closed |
| 129 | o4-miniOpenAI | -35.7% | Not reported | Closed |
| 130 | GLM-4.7Z.AI | -36.4% | Not reported | Open |
| 131 | Grok Code Fast 1xAI | -37.1% | Not reported | Closed |
| 132 | Step 3.7 FlashStepFun | -37.3% | Not reported | Open |
| 133 | Mistral Medium 3.5 128BMistral | -38.7% | Not reported | Open |
| 134 | MiniMax M2.5MiniMax | -38.9% | Not reported | Closed |
| 135 | GPT-4.1OpenAI | -39.6% | Not reported | Closed |
| 136 | Mistral Large 3Mistral | -39.6% | Not reported | Closed |
| 137 | DeepSeek V3 0324DeepSeek | -40.7% | Not reported | Open |
| 138 | DeepSeek V3DeepSeek | -41.6% | Not reported | Open |
| 139 | Gemini 2.0 FlashGoogle | -42.5% | Not reported | Closed |
| 140 | Gemini 2.5 FlashGoogle | -42.6% | Not reported | Closed |
| 141 | DeepSeek V3.1DeepSeek | -42.7% | Not reported | Open |
| 142 | Qwen3 MaxAlibaba | -43.5% | Not reported | Closed |
| 143 | Qwen3.5-27BAlibaba | -44.0% | Not reported | Open |
| 144 | Step 3.5 FlashStepFun | -44.2% | Not reported | Open |
| 145 | Magistral SmallMistral | -44.4% | Not reported | Open |
| 146 | DeepSeek V3.2DeepSeek | -46.9% | Not reported | Open |
| 147 | Nova ProAmazon | -47.7% | Not reported | Closed |
| 148 | Qwen3.5-35B-A3BAlibaba | -48.1% | Not reported | Open |
| 149 | MiMo-V2-FlashXiaomi | -48.4% | Not reported | Open |
| 150 | Gemma 4 31BGoogle | -48.5% | Not reported | Open |
| 151 | MiniMax M1 80kMiniMax | -48.5% | Not reported | Closed |
| 152 | Claude 3 HaikuAnthropic | -48.6% | Not reported | Closed |
| 153 | North Mini CodeCohere | -49.8% | Not reported | Open |
| 154 | Mistral Small 3.1Mistral | -50.7% | Not reported | Open |
| 155 | Mercury 2Inception | -50.7% | Not reported | Closed |
| 156 | Devstral SmallMistral | -50.8% | Not reported | Open |
| 157 | Grok 4.1 FastxAI | -50.9% | Not reported | Closed |
| 158 | Solar Pro 3Upstage | -53.4% | Not reported | Closed |
| 159 | GPT-4.1 miniOpenAI | -53.6% | Not reported | Closed |
| 160 | GPT-4.1 nanoOpenAI | -57.6% | Not reported | Closed |
| 161 | K-ExaoneLG AI Research | -58.0% | Not reported | Closed |
| 162 | Gemini 2.5 Flash-LiteGoogle | -58.8% | Not reported | Closed |
| 163 | GLM-4.5-AirZ.AI | -61.5% | Not reported | Closed |
| 164 | Solar Pro 2Upstage | -62.0% | Not reported | Closed |
| 165 | GLM-4.7-FlashZ.AI | -62.6% | Not reported | Open |
| 166 | Ling 2.6 FlashInclusionAI | -66.1% | Not reported | Open |
| 167 | Gemma 3 27BGoogle | -67.2% | Not reported | Open |
Among the reported AA-Omniscience Index rows, Gemini 4 Carbon is first at 53.5%. The third row is 7.2 points behind. The broader top-10 range is 22.9 points, so the table still separates the published systems.
167 models have been evaluated on AA-Omniscience Index. The benchmark falls in the Knowledge category. 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.
Freshness and 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.
Questions
What does AA-Omniscience Index measure?
A display-only Artificial Analysis factual knowledge index.
Which model scores highest on AA-Omniscience Index?
Gemini 4 Carbon by Google currently leads with a score of 53.5% on AA-Omniscience Index.
How many models are evaluated on AA-Omniscience Index?
167 AI models have published results on AA-Omniscience Index in the BenchLM catalog.
Compare top models on AA-Omniscience Index
Last updated: November 7, 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 issueJoin 6,100+ subscribers.
One email each week. Unsubscribe anytime.