NICE - Observational Imitation: leaderboard
Metric: Ranking accuracy (%) on the observational imitation dimension on NICE (137 human-written items in Chinese social contexts, each a ranking of an optimal, a suboptimal and a boundary-violating response), an item counts only when the full ranking matches the gold order, first-party APIs at temperature 0, three runs pooled; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 5 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 72.2 |
| 2 | Claude Opus 4.7 | 69.4 |
| 3 | DeepSeek V4 Pro | 61.1 |
| 4 | Qwen 3.6 Plus | 61.1 |
| 5 | GPT-5.5 | 55.6 |
Interactive version: theaggregate.ai/benchmark?slug=nice-observational-imitation · How It Works · Data refreshed daily, snapshot 2026-10-07.