Natural Questions
Natural Questions is a question answering dataset featuring real anonymized queries issued to Google search engine. It contains 307,373 training examples where annotators provide long answers (passages) and short answers (entities) from Wikipedia pages, or mark them as unanswerable.
Gemma 2 27B from Google currently leads the Natural Questions leaderboard with a score of 0.345 across 7 evaluated AI models.
What Natural Questions measures
Natural Questions is a text benchmark that evaluates large language models on general, reasoning, and search tasks. LLM Stats tracks 7 models on this benchmark, with a maximum possible score of 1. Current average across reported models is 0.2, with the leader reaching 0.3.
Compare leaders on the best AI for general, best AI for reasoning and best AI for search leaderboards.
Publication
- Paper
- A BERT Baseline for the Natural Questions
- Authors
- Chris Alberti, Kenton Lee, Michael Collins
- Published
- arXiv
- 1901.08634
Abstract
This technical note describes a new baseline for the Natural Questions. Our model is based on BERT and reduces the gap between the model F1 scores reported in the original dataset paper and the human upper bound by 30% and 50% relative for the long and short answer tasks respectively. This baseline has been submitted to the official NQ leaderboard at ai.google.com/research/NaturalQuestions. Code, preprocessed data and pretrained model are available at https://github.com/google-research/language/tree/master/language/question_answering/bert_joint.
Gemma 2 27B leads with 34.5%, followed by 
Mistral NeMo Instruct at 31.2% and Gemma 2 9B at 29.2%.
Progress Over Time
Interactive timeline showing model performance evolution on Natural Questions
Natural Questions Leaderboard
| Context | Cost | License | ||||
|---|---|---|---|---|---|---|
| 1 | Google | 27B | — | — | ||
| 2 | Mistral AI | 12B | — | — | ||
| 3 | Google | 9B | — | — | ||
| 4 | 2B | — | — | |||
| 4 | Google | 8B | — | — | ||
| 6 | 2B | — | — | |||
| 6 | Google | 8B | — | — |
FAQ
Common questions about Natural Questions.
More evaluations to explore
Related benchmarks in the same category
A challenging dataset of 448 multiple-choice questions written by domain experts in biology, physics, and chemistry. Questions are Google-proof and extremely difficult, with PhD experts reaching 65% accuracy.
A more robust and challenging multi-task language understanding benchmark that extends MMLU by expanding multiple-choice options from 4 to 10, eliminating trivial questions, and focusing on reasoning-intensive tasks. Features over 12,000 curated questions across 14 domains and causes a 16-33% accuracy drop compared to original MMLU.
All 30 problems from the 2025 American Invitational Mathematics Examination (AIME I and AIME II), testing olympiad-level mathematical reasoning with integer answers from 000-999. Used as an AI benchmark to evaluate large language models' ability to solve complex mathematical problems requiring multi-step logical deductions and structured symbolic reasoning.
Massive Multitask Language Understanding benchmark testing knowledge across 57 diverse subjects including STEM, humanities, social sciences, and professional domains
A verified subset of 500 software engineering problems from real GitHub issues, validated by human annotators for evaluating language models' ability to resolve real-world coding issues by generating patches for Python codebases.
Humanity's Last Exam (HLE) is a multi-modal academic benchmark with 2,500 questions across mathematics, humanities, and natural sciences, designed to test LLM capabilities at the frontier of human knowledge with unambiguous, verifiable solutions