PaperMind - Cross-Source Evidence Reasoning: leaderboard
Metric: Judge score for answering research questions that need evidence from the paper PDF, its cited references and external sources, as a smolagents ReAct agent with tools; rated by a GPT-4o judge on a 5-point scale (1-5) against the reference, averaged over the seven scientific domains; higher is better. Source: arxiv.org. Saturation forecast: Around January 2028. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 2.99 |
| 2 | GPT-4o Mini | 2.71 |
| 3 | Claude 3 Haiku | 2.49 |
| 4 | Qwen 3 VL 4B Instruct | 2.45 |
| 5 | Claude 3.5 Sonnet | 2.3 |
| 6 | Gemma 3 4B (IT) | 1.81 |
Interactive version: theaggregate.ai/benchmark?slug=papermind-cross-source-evidence-reasoning · How It Works · Data refreshed daily, snapshot 2026-10-07.