VSI-Super-Wild: leaderboard
VSI-Super-Wild (ECCV 2026) probes spatial supersensing in genuinely long in-the-wild videos — up to 4+ hours — with human-verified questions on motion, ordering, and continuous counting; scored by overall accuracy.
Metric: Overall Accuracy (%). Source: vsi-super-wild.github.io. Saturation forecast: Around 2031. 14 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 44.36 | #58 |
| 2 | GPT-5.4 | 34.52 | #79 |
| 3 | Qwen 3.5 9B | 33.87 | #379 |
| 4 | Qwen 2.5 VL 7B | 30.98 | #643 |
| 5 | Qwen 2 VL 7B | 26.67 | #750 |
| 6 | Gemini 3.1 Flash Lite | 23.85 | #191 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=vsi-super-wild · How It Works · Data refreshed daily, snapshot 2026-10-20.