VistaQA - Answer Accuracy: leaderboard
Metric: Overall text answer accuracy (%), on the 1,157 VistaQA samples (six task types, six visual domains, 314 hallucination samples whose queried entity is absent), zero-shot with standardized output-format instructions; the answer is judged correct or not by Qwen 2.5-14B and the evidence masks are matched to the reference masks by Hungarian IoU; higher is better. Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | SAM3 + Gemini 3 (VistaQA checkpoint unspecified) | 62.32 |
| 2 | SAM3 + GPT-5.4 (Thinking) | 61.02 |
| 3 | SAM3 + Qwen3-VL-32B-Instruct | 57.65 |
| 4 | SAM3 + GPT-5.4 | 53.5 |
| 5 | SAM3 + Qwen3-VL-4B-Instruct | 53.15 |
| 6 | R-Sa2VA-Qwen3VL-4B-RL | 36.3 |
| 7 | Sa2VA-8B | 29.04 |
| 8 | UniPixel-7B | 21.26 |
| 9 | TreeVGR-7B | 16.08 |
| 10 | LISA-7B | 7.26 |
Interactive version: theaggregate.ai/benchmark?slug=vistaqa-answer-accuracy · How It Works · Data refreshed daily, snapshot 2026-10-07.