Deep-Research Citation Benchmark - Source Relevance: leaderboard
Metric: Cohen kappa (-1 to 1) between the LLM judge and human-reviewed gold labels on the source relevance rubric criterion of 624 attribution-citation pairs parsed from an adversarial 25-domain deep-research report (scores binarized at 0.5); higher is better. Source: arxiv.org. Saturation forecast: Around February 2028. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 Mini | 0.64 |
| 2 | Claude Opus 4.6 | 0.55 |
| 3 | GPT-5.4 Mini | 0.54 |
| 4 | GPT-OSS-120B | 0.52 |
| 5 | Gemini 3.1 Pro (Preview) | 0.46 |
| 6 | Gemini 3.1 Flash Lite (Preview) | 0.46 |
| 7 | Claude Haiku 4.5 | 0.34 |
| 8 | Claude Sonnet 4.6 | 0.32 |
Interactive version: theaggregate.ai/benchmark?slug=deep-research-citation-benchmark-source-relevance · How It Works · Data refreshed daily, snapshot 2026-09-29.