evals.report
BenchmarksLabsCompareRun guidesIn the wild
BenchmarksReasoning

AA-Omniscience v1.1: Knowledge and Hallucination Benchmark

Artificial Analysis' revised factual-knowledge and hallucination benchmark: 3,000 private questions across six major domains, down from 6,000 in v1. Saturated and public-set questions were removed, new questions were added, and ambiguous or erroneous ones were revised. Questions are drawn from authoritative academic and industry sources by an LLM-based question-generation agent. The Net Score (-100 to 100) rewards correct answers, penalizes hallucinated ones, and does not penalize abstaining.

ReasoningAA-Omniscience Net ScoreHigher is better

What is AA-Omniscience v1.1: Knowledge and Hallucination Benchmark?

Artificial Analysis' revised factual-knowledge and hallucination benchmark: 3,000 private questions across six major domains, down from 6,000 in v1. Saturated and public-set questions were removed, new questions were added, and ambiguous or erroneous ones were revised. Questions are drawn from authoritative academic and industry sources by an LLM-based question-generation agent. The Net Score (-100 to 100) rewards correct answers, penalizes hallucinated ones, and does not penalize abstaining. evals.report tracks reported AA-Omniscience v1.1: Knowledge and Hallucination Benchmark scores with the model, source, status, date, and run caveats attached — official leaderboard scores, vendor-reported launches, and clearly labeled community runs.

Top reported AA-Omniscience v1.1: Knowledge and Hallucination Benchmark score: Claude Fable 5.2 — 67 (AA-Omniscience Net Score).

ModelLabScore↓Source modelStatusDate
Claude Fable 5.2Anthropic67Claude Fable 5.2 (Max, Default Fallback)OfficialNov 9, 2026Details
Gemini 4 CarbonGoogle DeepMind54Gemini 4 Carbon (High)OfficialNov 10, 2026Details
Claude Opus 5.5Anthropic48Claude Opus 5.5 (Max, Default Fallback)OfficialSep 22, 2026Details
Gemini 4 HydrogenGoogle DeepMind46Gemini 4 Hydrogen (Xhigh)OfficialNov 12, 2026Details
Gemini 4 ArgonGoogle DeepMind46Gemini 4 Argon (High)OfficialOct 13, 2026Details
GPT-6 AstraOpenAI45GPT-6 Astra (High)OfficialSep 3, 2026Details
GPT-6.1 SolOpenAI42GPT-6.1 Sol (Xhigh)OfficialSep 29, 2026Details
Grok 4.7xAI31Grok 4.7 (Xhigh)OfficialSep 21, 2026Details
Muse Wonder FullMeta31Muse Wonder Full (Max)OfficialOct 12, 2026Details
Claude Sonnet 5.5Anthropic30Claude Sonnet 5.5 (Max, Default Fallback)OfficialSep 28, 2026Details
Kimi K3.1 ProOpenMoonshot AI30Kimi K3.1 ProOfficialOct 23, 2026Details
GPT-6.1 TerraOpenAI30GPT-6.1 Terra (Max)OfficialOct 26, 2026Details
Muse Spark OpenOpenMeta26Muse Spark Open (Xhigh)OfficialOct 15, 2026Details
Muse WonderMeta25Muse Wonder (Max)OfficialOct 12, 2026Details
Gemini 3.8 FlashGoogle DeepMind24Gemini 3.8 Flash (Medium)OfficialSep 2, 2026Details
Step 5 PreviewStepFun16Step 5 PreviewOfficialSep 18, 2026Details
Inkling 2 SmallOpenThinking Machines Lab16Inkling 2 SmallOfficialNov 5, 2026Details
Claude Haiku 5.5Anthropic12Claude Haiku 5.5 (Max)OfficialOct 7, 2026Details
Qwen3.8 Max (0902)Alibaba / Qwen10Qwen3.8 Max (0902)OfficialSep 2, 2026Details
Qwen3.8 MaxAlibaba / Qwen8Qwen3.8 2.4T A95BOfficialAug 3, 2026Details
Muse EmberMeta8Muse Ember (Xhigh)OfficialNov 9, 2026Details
Gemini 3.5 Flash-LiteGoogle DeepMind7Gemini 3.5 Flash-LiteOfficialJul 21, 2026Details
GLM-5.3OpenZ.ai6GLM-5.3 (Max)OfficialAug 14, 2026Details
Hy4Tencent5Tencent Hy4OfficialNov 4, 2026Details
Mistral Large 4OpenMistral AI5Mistral Large 4 (High)OfficialOct 29, 2026Details
NVIDIA Nemotron 3 UltraOpenNVIDIA3Nemotron 3 Ultra 550B A55B (Reasoning)OfficialJun 4, 2026Details
Beam 1.0OpenReflection AI2Beam 1.0OfficialOct 16, 2026Details
DeepSeek V4.1 Flash 1017OpenDeepSeek2DeepSeek V4.1 Flash 1017 (Non-reasoning)OfficialOct 17, 2026Details
JT-4.1 Flash 236B A21BChina Mobile2JT-4.1 Flash 236B A21BOfficialNov 11, 2026Details
GLM-5.3-FlashOpenZ.ai1GLM-5.3-FlashOfficialAug 26, 2026Details
Ling 3.1 Flash 1019OpeninclusionAI (Ant Group)0Ling 3.1 Flash 1019OfficialOct 27, 2026Details
InklingOpenThinking Machines Lab-2Inkling (Xhigh)OfficialJul 15, 2026Details
NVIDIA Nemotron 3.5 Super VL 120B-A12BOpenNVIDIA-3Nemotron 3.5 Super VLOfficialOct 15, 2026Details
Nanbeige4.3-33BOpenNanbeige LLM Lab-4Nanbeige4.3-33BOfficialOct 25, 2026Details
MiMo-V2.6-ProOpenXiaomi-4MiMo-V2.6-ProOfficialSep 21, 2026Details
K2 Horizon 375B A23BOpenMBZUAI-5K2 Horizon 375B A23BOfficialSep 3, 2026Details
MiniMax M3OpenMiniMax-6MiniMax-M3OfficialJun 1, 2026Details
Qwen3.8-27BOpenAlibaba / Qwen-6Qwen3.8 27B (Xhigh)OfficialAug 5, 2026Details
DeepSeek V4 Pro 0813OpenDeepSeek-6DeepSeek V4 Pro 0813 (Max)OfficialAug 13, 2026Details
MiniMax M3.1 FlashOpenMiniMax-6MiniMax-M3.1-FlashOfficialOct 19, 2026Details
GPT-6 LunaOpenAI-8GPT-6 Luna (Max)OfficialSep 22, 2026Details
Inkling SmallOpenThinking Machines Lab-9Inkling SmallOfficialJul 30, 2026Details
MiMo-V2.6-FlashOpenXiaomi-11MiMo-V2.6-FlashOfficialSep 21, 2026Details
Agnes 2.5 Pro ThetaOpenSapiens AI-11Agnes 2.5 Pro ThetaOfficialNov 7, 2026Details
Solar Mini 4Upstage-12Solar Mini 4OfficialSep 22, 2026Details
NVIDIA Nemotron 3.5 Lightning 30B-A3BOpenNVIDIA-16Nemotron 3.5 LightningOfficialAug 11, 2026Details
Muse Glimmer 30BOpenMeta-32Muse Glimmer (High)OfficialAug 10, 2026Details
Mistral Medium 3.5OpenMistral AI-39Mistral Medium 3.5OfficialApr 28, 2026Details
GPT-OSS-120BOpenOpenAI-46gpt-oss-120b (High)OfficialAug 5, 2025Details

Each row reports the model’s AA-Omniscience Net Score on AA-Omniscience v1.1: Knowledge and Hallucination Benchmark. Click a row for the full run context.