BehaviorBench - Multi-Round Game Prediction (Distribution): leaderboard

Metric: Given the game rules and the earlier rounds of each subject, predict the next moves, on MobLab economic-game records of human players (dictator, ultimatum, trust, public goods, bomb risk, beauty contest and push/pull games): Wasserstein distance (0-100) between the model-predicted and the observed human choice distributions, with choices normalized to 0-100 and averaged over games; lower is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 18 models tracked.

Top models

#ModelScore
1GPT-4.16.9
2DeepSeek V3.210.1
3Claude Sonnet 4.610.6
4Claude Opus 4.611.7
5Gemini 3.1 Pro (Preview)12.5
6Claude Haiku 4.512.5
7Qwen 3 4B13.1
8GPT-5.4 (High)15.2
9Llama 3.3 70B Instruct15.3
10GPT-5.4 Mini (High)17.6
11Gemini 3.1 Flash Lite (Preview)18.6

Interactive version: theaggregate.ai/benchmark?slug=behaviorbench-multi-round-game-prediction-distribution · How It Works · Data refreshed daily, snapshot 2026-09-29.