VANTAGE-Bench - Event Verification: leaderboard
Metric: Event Verification (%). Source: huggingface.co. 17 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.6 Flash | 82 |
| 2 | GPT-5.6 Sol | 75.9 |
| 3 | Gemini 3.5 Flash Lite | 70.41 |
| 4 | Gemini 3.1 Pro (Preview) | 69.83 |
| 5 | Qwen 3 VL 32B Instruct | 60.04 |
| 6 | Qwen 3 VL 8B Instruct | 59.39 |
| 7 | Qwen 3.5 27B | 55.34 |
| 8 | Gemini 3.1 Flash Lite | 54.83 |
| 9 | Qwen 3.5 9B | 46.78 |
| 10 | Qwen 3 VL 2B Instruct | 44.73 |
| 11 | gemma-4-E2B-it | 27.65 |
Interactive version: theaggregate.ai/benchmark?slug=vantage-bench-event-verification · How It Works · Data refreshed daily, snapshot 2026-09-05.