ESI-Bench - Physical Contact: leaderboard
Metric: Accuracy (%; Physical Contact questions in the Spatial Relations category, active multi-view exploration). Source: arxiv.org. Saturation forecast: Around December 2026. 2 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 | 64.2 |
Interactive version: theaggregate.ai/benchmark?slug=esi-bench-physical-contact · How It Works · Data refreshed daily, snapshot 2026-09-26.