MechVQA - Anomaly Detection: leaderboard
Metric: Accuracy (%) on the Anomaly Detection subtask (Judging capability), MechVQA test split (drawing-level 8:1:1 split of 20,778 questions on 3,281 mechanical drawings), answers judged against the reference by three LLM judges (GPT-OSS-120B, DeepSeek-V3.2, Kimi-k2); higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 15 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 78.37 |
| 2 | Qwen 3 VL 32B Instruct | 75.92 |
| 3 | GLM-4.6V | 74.29 |
| 4 | GPT-5 | 71.02 |
| 5 | Claude Sonnet 4.5 | 64.9 |
| 6 | Qwen 3 VL 30B A3B Instruct | 64.08 |
| 7 | Qwen 3 VL 4B Instruct | 62.86 |
| 8 | GPT-4o | 53.06 |
| 9 | GPT-4o Mini | 35.1 |
| 10 | Gemma 3 27B (IT) | 24.9 |
| 11 | Llama 3.2 11B Instruct | 17.55 |
Interactive version: theaggregate.ai/benchmark?slug=mechvqa-anomaly-detection · How It Works · Data refreshed daily, snapshot 2026-10-07.