SciImpact - Patent: leaderboard

SciImpact pairs that differ in how often patents cite the paper, from titles and abstracts; the model sees two artifacts from the same field and picks the one with higher impact; pairwise accuracy in percent, chance 50, higher is better.

Metric: Pairwise prediction accuracy (%, times 100) on patent-citation pairs, over SciImpact contrastive pairs (215,928 pairs over 19 fields, built from citation counts, best paper awards and Nobel prizes, patent citations, media attention, GitHub stars and Hugging Face downloads with impact thresholds): the model sees two same-field artifacts (title and abstract, README, dataset card or model card, truncated to 1,000 words) and names the one with higher future impact in a two-option forced choice parsed by exact match; the order is balanced so chance is 50%; higher is better. Source: arxiv.org. Saturation forecast: Around 2030. 11 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1O4 Mini60.4#368
2Claude Haiku 4.559.6#583
3GPT-4.1 Mini59.2#789
4Qwen 2.5 14B58.6#1811
5Qwen 2.5 7B55.7#2929
6Qwen 3 4B54.9#2348
7Ministral 3 3B53.6#3504
8Llama 3.1 8B53.4#5196
9Llama 3.2 3B53.4#7877
10Llama 3 8B53.4#5900
11nemotron-3-nano-30B-a3B50#1946

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=sciimpact-patent · How It Works · Data refreshed daily, snapshot 2026-11-21.