AA-Omniscience: Knowledge and Hallucination Benchmark
A factuality and knowledge benchmark of 6,000 questions across 42 economically relevant topics in six domains, scoring models on the AA-Omniscience Index (-100 to 100) that rewards correct answers, penalizes hallucinations, and applies no penalty for abstaining.
What this benchmark measures
A factuality and knowledge benchmark of 6,000 questions across 42 economically relevant topics in six domains, scoring models on the AA-Omniscience Index (-100 to 100) that rewards correct answers, penalizes hallucinations, and applies no penalty for abstaining.
Rows on this page are sourced from public benchmark artifacts, leaderboard exports, or source-linked model reports. Each row keeps benchmark version, source model name, and available run details attached to the score.
The metric shown here is AA-Omniscience Index. It should be interpreted within AA-Omniscience: Knowledge and Hallucination Benchmark, not compared as part of a site-wide ranking.
Frequently asked
What is AA-Omniscience: Knowledge and Hallucination Benchmark?
A factuality and knowledge benchmark of 6,000 questions across 42 economically relevant topics in six domains, scoring models on the AA-Omniscience Index (-100 to 100) that rewards correct answers, penalizes hallucinations, and applies no penalty for abstaining. It is a reasoning benchmark measured by AA-Omniscience Index.
What does AA-Omniscience Index mean on AA-Omniscience: Knowledge and Hallucination Benchmark?
AA-Omniscience: Knowledge and Hallucination Benchmark reports AA-Omniscience Index; higher is better. Scores are shown only within AA-Omniscience: Knowledge and Hallucination Benchmark and are never averaged with other benchmarks.
What is the top reported AA-Omniscience: Knowledge and Hallucination Benchmark score?
Gemini 3.1 Pro Preview has the top reported score on AA-Omniscience: Knowledge and Hallucination Benchmark: 33 (AA-Omniscience Index).
Why do AA-Omniscience: Knowledge and Hallucination Benchmark scores differ across runs?
Harness, scaffold, reasoning effort, and prompt setup change results, so two runs of the same model can differ. evals.report keeps each score with its run context so the differences stay visible.
Does evals.report rank models across benchmarks?
No. AA-Omniscience: Knowledge and Hallucination Benchmark scores are shown within their own metric; evals.report never combines benchmarks into a composite ranking or a single "best model".