MechVQA - Structure Understanding: leaderboard

MechVQA (2026) tests multimodal LLMs on understanding mechanical engineering drawings, with 20,778 questions on 3,281 drawings scored as accuracy by three LLM judges against reference answers. This board covers the Structure Understanding subtask of the reasoning capability, as accuracy in percent on the test split; higher is better.

Metric: Accuracy (%) on the Structure Understanding subtask (Reasoning 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 February 2027. 15 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Qwen 3 VL Plus81.25#512
2GPT-579.17#178
3GLM-4.6V77.08#710
4Claude Sonnet 4.570.83#261
5GPT-4o64.58#751
6Qwen 3 VL 32B Instruct62.5#606
7Gemini 3 Pro (Preview)52.08#119
8MiniCPM-V-4.545.83#1441
9Qwen 3 VL 30B A3B Instruct43.75#895
10Qwen 3 VL 4B Instruct39.58#1332
11InternVL3.5-8B37.5#1604
12Gemma 3 27B (IT)31.25#1324
13GPT-4o Mini29.17#1667
14MiMo-VL-7B-SFT27.08#1089
15Llama 3.2 11B Instruct6.25#5174

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=mechvqa-structure-understanding · How It Works · Data refreshed daily, snapshot 2026-11-21.