BFI-Adapt: leaderboard
Metric: BFI-Adapt composite (%; mean over the 27 event-trait pairs with a definite human change direction of max(0, weighted kappa) x (2 DCR - 1) x an indicator that the dominant item-change direction matches the human prior; 100 personas simulate 11 life events and answer the BFI-44 before and after; non-reasoning mode). Source: arxiv.org. Saturation forecast: Around 2029. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash (Minimal) | 34.8 |
| 2 | GLM-4.6 (Non-reasoning) | 32.2 |
| 3 | Qwen 3 235B A22B (Non-reasoning) | 28.7 |
| 4 | Claude Haiku 4.5 | 22.5 |
| 5 | Kimi K2 0905 | 22.4 |
| 6 | Seed 2.0 Pro (Non-reasoning) | 21.8 |
| 7 | Claude Sonnet 4.6 | 21.2 |
| 8 | Qwen 3.5 9B (Non-reasoning) | 19.4 |
| 9 | GPT-4.1 Mini | 19.2 |
| 10 | GPT-5.3 | 17.8 |
| 11 | DeepSeek V4 Pro (Non-reasoning) | 17.6 |
| 12 | MiMo-V2.5-Pro (Non-reasoning) | 7.1 |
| 13 | Qwen 2.5 14B Instruct | 4.9 |
| 14 | InternLM3-8B-Instruct | 4.4 |
Interactive version: theaggregate.ai/benchmark?slug=bfi-adapt · How It Works · Data refreshed daily, snapshot 2026-09-26.