Scale-VQA - Localization: leaderboard
Metric: Accuracy (%; multiple-choice questions on paired 10x/40x/200x views of TCGA whole-slide images, Scale-VQA test split, 201 localization questions). Source: arxiv.org. Saturation forecast: Around September 2027. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.2 | 74.13 |
| 2 | Gemini 3 Flash | 71.64 |
| 3 | Qwen 2.5 VL 7B Instruct | 71.14 |
| 4 | Lingshu-7B | 62.69 |
Interactive version: theaggregate.ai/benchmark?slug=scale-vqa-localization · How It Works · Data refreshed daily, snapshot 2026-09-26.