BehaviorBench - Survey Response from Demographics: leaderboard
Metric: Accuracy (%) of predicting the five-point answer of a subject to one item from their demographics, on the Big Five personality survey responses (50 five-point items) of held-out subjects; chance 20; higher is better. Source: arxiv.org. Saturation forecast: Around 2036. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 30.5 |
| 2 | Gemini 3.1 Flash Lite (Preview) | 30.5 |
| 3 | Claude Opus 4.6 | 29.7 |
| 4 | GPT-5.4 (High) | 29.6 |
| 5 | GPT-4.1 | 29.2 |
| 6 | GPT-5.4 Mini (High) | 28.8 |
| 7 | Claude Sonnet 4.6 | 28.6 |
| 8 | Llama 3.3 70B Instruct | 28.4 |
| 9 | DeepSeek V3.2 | 27.2 |
| 10 | Claude Haiku 4.5 | 26 |
| 11 | Qwen 3 4B | 24.3 |
Interactive version: theaggregate.ai/benchmark?slug=behaviorbench-survey-response-from-demographics · How It Works · Data refreshed daily, snapshot 2026-09-29.