KDR-Bench - Main Conclusion Alignment: leaderboard
Metric: Main conclusion alignment (0-100): a 0-10 judge score of how well the report reaches the annotated main conclusion, times 10, on the 41 expert-level research questions of KDR-Bench in nine domains backed by 1,252 Statista tables, each report judged by DeepSeek-V3.2; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 Max | 64.2 |
| 2 | Deep Research | 63.5 |
| 3 | GLM-4.6 | 59.1 |
| 4 | Tongyi DeepResearch 30B A3B | 58.6 |
| 5 | MiniMax-M2 | 58.2 |
Interactive version: theaggregate.ai/benchmark?slug=kdr-bench-main-conclusion-alignment · How It Works · Data refreshed daily, snapshot 2026-10-07.