UltraVR - Anomaly Detection: leaderboard
Metric: Accuracy (%) on the anomaly detection domain: locate the one logically or structurally anomalous cell (row and column) in a composed grid of 100 MVTec LOCO cells, direct answering (S0) with the original ultra-resolution image under each model default preprocessing (V2); four-choice questions except anomaly detection, which asks for the grid coordinate of the one anomalous cell out of 100; higher is better. Source: arxiv.org. Saturation forecast: Around December 2027. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 30.8 |
| 2 | Gemini 3.1 Pro (Preview) | 30 |
| 3 | GPT-5.4 | 20.8 |
| 4 | Qwen 3.5 27B | 20.8 |
| 5 | Gemma 4 26B | 9.2 |
| 6 | GLM-4.5V | 5.7 |
Interactive version: theaggregate.ai/benchmark?slug=ultravr-anomaly-detection · How It Works · Data refreshed daily, snapshot 2026-09-29.