SSMNBench - Single-View Sufficiency: leaderboard
Metric: Mean accuracy (%) over the five single-view-sufficiency tasks (fine-grained action, human-human contact, human-object contact, relative self-pose, anatomical joint), where one view suffices; four-option multiple choice on dense, occlusion-heavy multi-view scenes of people, given exactly the annotated necessary views (the +0 setting); 300 questions per task, images resized to 1920x1080, single-letter answers. Source: arxiv.org. Saturation forecast: Around August 2028. 16 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.2 | 47.5 |
| 2 | Gemini 2.5 Pro | 44.6 |
| 3 | Gemini 2.5 Flash | 42.7 |
| 4 | Qwen 2.5 VL 72B Instruct | 42.7 |
| 5 | Qwen 2.5 VL 32B Instruct | 40.3 |
| 6 | InternVL3-78B | 36.2 |
| 7 | Qwen 2.5 VL 7B Instruct | 31.3 |
Interactive version: theaggregate.ai/benchmark?slug=ssmnbench-single-view-sufficiency · How It Works · Data refreshed daily, snapshot 2026-09-29.