DrawingVQA - Perceptual Understanding: leaderboard
Metric: Accuracy (%; reasoning depth R1, 31 perceptual-understanding questions of the 92 expert-curated questions on 33 Issued for Construction drawings, multiple choice and open-ended). Source: arxiv.org. Saturation forecast: Around December 2026. 19 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 80.6 |
| 2 | Gemini 3 Pro (Preview) | 80.6 |
| 3 | Gemini 2.5 Flash | 77.4 |
| 4 | Claude Sonnet 4.5 | 64.5 |
| 5 | O3 (2025-04-16) | 64.5 |
| 6 | GPT-4o (2024-08-06) | 61.3 |
| 7 | Llama 4 Scout Instruct | 58.1 |
| 8 | InternVL3.5-8B | 58.1 |
| 9 | Claude Haiku 4.5 | 51.6 |
| 10 | Qwen 3 VL 8B Instruct | 51.6 |
| 11 | Qwen 3 VL 30B A3B Instruct | 51.6 |
| 12 | Qwen 3 VL 32B Instruct | 48.4 |
| 13 | Llama 3.2 11B Instruct | 45.2 |
| 14 | Phi-4 Multimodal Instruct | 41.9 |
Interactive version: theaggregate.ai/benchmark?slug=drawingvqa-perceptual-understanding · How It Works · Data refreshed daily, snapshot 2026-09-29.