360Bench (ERP) - Projection-Distorted Perception: leaderboard
Metric: Projection-distorted perception accuracy (%), mean of recognition and counting of objects distorted or split by the projection on 360Bench's 1,532 human-written four-option questions on 643 high-resolution (7K) 360-degree images, each scored under four cyclic permutations of the options (6,128 evaluations), greedy decoding, where the image is given in equirectangular projection (ERP); higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o | 46.5 | #333 |
| 2 | Gemini 2.5 Flash | 39.2 | #237 |
| 3 | Qwen 2.5 VL 7B Instruct | 39.1 | #643 |
| 4 | Gemini 2.5 Pro | 37.9 | #145 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=360bench-erp-projection-distorted-perception · How It Works · Data refreshed daily, snapshot 2026-10-11.