MMR-AD - MVTec AD Detection: leaderboard
Metric: Anomaly detection accuracy (%): share of MVTec AD test images (from MMR-AD) correctly judged normal or anomalous, zero-shot, the model sees a spatially aligned normal reference image and the test image with the MMR-AD instruction template; dataset-level average over the subdataset's product classes; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | InternVL3-38B | 83.8 |
| 2 | O4 Mini | 81.3 |
| 3 | Gemini 2.5 Pro | 79.4 |
| 4 | GPT-5 | 78.7 |
| 5 | InternVL3-8B | 78.4 |
| 6 | Llama 4 Maverick | 76.1 |
| 7 | Qwen 2.5 VL 7B | 75 |
| 8 | GPT-4o | 68.9 |
Interactive version: theaggregate.ai/benchmark?slug=mmr-ad-mvtec-ad-detection · How It Works · Data refreshed daily, snapshot 2026-10-07.