ARC

A benchmark for evaluating language models across diverse subjects spanning humanities, social sciences, STEM, and professional knowledge.

A benchmark that evaluates the ability of AI models to answer challenging graduate-level multiple-choice questions across scientific domains, measuring expert-level reasoning, scientific knowledge, and problem-solving performance.

A benchmark that evaluates the commonsense reasoning and natural language understanding capabilities of AI models by requiring them to select the most plausible continuation of a given context from multiple candidate endings.

A benchmark measuring how truthfully language models answer questions and avoid mimicking human falsehoods.

A benchmark that evaluates the ability of AI models to solve grade school mathematical word problems requiring multi-step reasoning, arithmetic, and logical problem-solving.