CRAG
CRAG (Comprehensive RAG Benchmark) is a factual question answering benchmark consisting of 4,409 question-answer pairs across 5 domains (finance, sports, music, movie, open domain) and 8 question categories. The benchmark includes mock APIs to simulate web and Knowledge Graph search, designed to represent the diverse and dynamic nature of real-world QA tasks with temporal dynamism ranging from years to seconds. It evaluates retrieval-augmented generation systems for trustworthy question answering.
Nova Pro from Amazon currently leads the CRAG leaderboard with a score of 0.503 across 3 evaluated AI models.
Nova Pro leads with 50.3%, followed by Nova Lite at 43.8% and Nova Micro at 43.1%.
Progress Over Time
Interactive timeline showing model performance evolution on CRAG
CRAG Leaderboard
| Context | Cost | License | ||||
|---|---|---|---|---|---|---|
| 1 | Amazon | — | — | — | ||
| 2 | Amazon | — | — | — | ||
| 3 | Amazon | — | — | — |
FAQ
Common questions about CRAG.
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