LitReviewBench - Research Suggestions: leaderboard

Metric: Expert-preference rating (Elo scale, Bradley-Terry aggregation of about 3,000 anonymized pairwise battles judged by matched domain experts on D4 quality of identified gaps and future directions; literature review drafts on AI topics; the human-draft baseline is in the rating pool). Source: arxiv.org. Saturation forecast: Around March 2027. 9 models tracked.

Top models

#ModelScore
1Sonar Deep Research1322.7
2Claude Opus 4.51099.9
3Qwen 3 235B A22B877.4
4Grok 4778.2
5GLM-4.6608.5
6Gemini 2.5 Pro469.8

Interactive version: theaggregate.ai/benchmark?slug=litreviewbench-research-suggestions · How It Works · Data refreshed daily, snapshot 2026-09-26.