GapEval - World Knowledge (Understanding): leaderboard
Metric: Accuracy (%; world-knowledge questions about entities shown in images; text answers to GapEval's bidirectional questions, judged correct or incorrect by GPT-5-mini against reference answers, mean of ten sampled runs). Source: arxiv.org. Saturation forecast: Estimated already saturated. 15 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Flash | 97.96 |
| 2 | GPT-5 Mini | 97.28 |
| 3 | Qwen 3 VL 235B A22B | 96.6 |
| 4 | GPT-5 | 96.52 |
| 5 | Qwen 3 VL 8B | 94.3 |
Interactive version: theaggregate.ai/benchmark?slug=gapeval-world-knowledge-understanding · How It Works · Data refreshed daily, snapshot 2026-09-26.