MiGUE-Bench - Event Detection: leaderboard
Metric: Micro-F1 (%) on 330 event-trigger detection questions (multiple-choice questions with LLM-built distractors over news documents, several correct options); higher is better. Source: arxiv.org. Saturation forecast: Around December 2026. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.5 | 85.52 |
| 2 | GPT-5.2 Pro | 83.98 |
| 3 | DeepSeek V3.2 | 81.68 |
| 4 | GLM-4.7 | 81.33 |
| 5 | Gemini 3 Pro | 71.51 |
| 6 | Qwen 3 Max | 65.43 |
| 7 | Qwen 3 30B A3B | 62.98 |
| 8 | Qwen 3 8B | 61.38 |
| 9 | Qwen 3 235B A22B | 61.15 |
| 10 | Claude Haiku 4.5 | 61.01 |
| 11 | Kimi K2 | 58.94 |
Interactive version: theaggregate.ai/benchmark?slug=migue-bench-event-detection · How It Works · Data refreshed daily, snapshot 2026-09-29.