AGC-Bench: Measuring Artificial General Creativity
2026-07-01 • Computation and Language
Computation and Language
AI summaryⓘ
The authors created AGC-Bench, a big test to measure creativity in AI models across different areas like writing, science, and humor. They fixed judge bias by training a special AI judge called AGC-Judge to score creativity fairly. Their tests showed AI models have varying creativity in different tasks and discovered a single creativity factor, like intelligence, that mostly explains their creative abilities. They also found that telling AI to "be creative" helps more than just making it reason better. Humans still outperform AI in creativity on some tasks, and the authors provide all tools and data publicly for others to use.
creativitylarge language models (LLMs)AGC-BenchJudge Response Theoryfactor analysispsychometricsbenchmarkinggeneral intelligenceAI evaluationprompting
Authors
Roger Beaty, Vijeta Deshpande, Clin K. Y. Lai, Anna Attuch, Namrata Shivagunde, Swastik Roy, Rajkumar Pujari, Paul V. DiStefano, Sherin Muckatira, Claire E. Stevenson, Mikhail Gronas, Anna Rumshisky
Abstract
Creativity research has debated whether creativity is domain-specific (e.g., visual, writing, science), and if it is psychometrically separable from general intelligence. Both questions now apply to LLMs, but a unified benchmark of AI creativity remains elusive. We introduce AGC-Bench, an artificial general creativity benchmark built from a systematic review of the AI creativity literature (3,101 papers screened, 497 benchmarks identified), paired with an agentic harness that converts idiosyncratic codebases into HELM-standardized benchmarks. The first release covers 78 datasets spanning brainstorming, problem solving, STEM, narrative, figurative language, and humor. To address bias in LLM-as-judge, we apply Judge Response Theory -- a psychometric calibration of judge leniency/severity; we then fine-tune Qwen3-30B on the bias-corrected ratings of three frontier LLMs to produce AGC-Judge, an open-weight model that robustly scores new creativity benchmarks it was not trained on. Results reveal frontier models at the top of the AGC-Bench leaderboard, with open models close behind. LLMs show different creative strengths, ranking higher on some domains (e.g., writing) than others (e.g., scientific ideation). Extensive experiments yield three main findings. First, applying factor analysis across 83 LLMs, we recover a single creativity factor 'c', analogous to the 'g' factor of general intelligence, that explains 81.5% of variance, related to but separable from general knowledge/reasoning. Second, we show that prompting models to "be creative" boosts their performance far more than enabling reasoning, evidence that the benchmark tracks creativity over general ability. Third, on a human-matched subset, we find the top human still leads the top LLM on creativity. We release AGC-Bench with a public leaderboard, AGC-Judge, and human data as open infrastructure for measuring AI creativity at scale.