MiGUE-Bench - Intra-Document Temporal Relations: leaderboard
Metric: Accuracy (%) on 300 single-document temporal relation multiple-choice questions with LLM-built distractors over news documents; higher is better. Source: arxiv.org. Saturation forecast: Around December 2026. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.5 | 71 |
| 2 | Qwen 3 Max | 69.67 |
| 3 | GLM-4.7 | 69.33 |
| 4 | Gemini 3 Pro | 68.67 |
| 5 | Claude Haiku 4.5 | 67.67 |
| 6 | DeepSeek V3.2 | 66 |
| 7 | GPT-5.2 Pro | 66 |
| 8 | Kimi K2 | 59 |
| 9 | Qwen 3 235B A22B | 56 |
| 10 | Qwen 3 30B A3B | 41 |
| 11 | Qwen 3 8B | 40 |
Interactive version: theaggregate.ai/benchmark?slug=migue-bench-intra-document-temporal-relations · How It Works · Data refreshed daily, snapshot 2026-09-29.