VNU-Bench - Cross-source Evidence Integration: leaderboard
Metric: Accuracy (%; question type T6, 4-option multiple choice). Source: arxiv.org. Saturation forecast: Around 2029. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 VL 8B | 60.63 |
| 2 | Claude Sonnet 4.5 | 60.08 |
| 3 | Gemini 2.5 Pro | 60.02 |
| 4 | Qwen 2.5 VL 7B Instruct | 49.22 |
| 5 | InternVL3-8B | 49.06 |
| 6 | Gemini 2.5 Flash | 45.96 |
Interactive version: theaggregate.ai/benchmark?slug=vnu-bench-cross-source-evidence-integration · How It Works · Data refreshed daily, snapshot 2026-09-25.