BehaviorBench - Cross-Game Prediction (Distribution): leaderboard
Metric: Given the first-round moves of each subject in other games, predict the first-round moves in a target game, 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: Around December 2026. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 10.3 |
| 2 | DeepSeek V3.2 | 12.3 |
| 3 | GPT-5.4 (High) | 14 |
| 4 | Llama 3.3 70B Instruct | 14.9 |
| 5 | GPT-4.1 | 16.3 |
| 6 | Claude Haiku 4.5 | 17.3 |
| 7 | Gemini 3.1 Flash Lite (Preview) | 17.3 |
| 8 | GPT-5.4 Mini (High) | 18 |
| 9 | Claude Sonnet 4.6 | 19.3 |
| 10 | Claude Opus 4.6 | 20.1 |
| 11 | Qwen 3 4B | 20.1 |
Interactive version: theaggregate.ai/benchmark?slug=behaviorbench-cross-game-prediction-distribution · How It Works · Data refreshed daily, snapshot 2026-09-29.