SciImpact - Physics: leaderboard

SciImpact impact-prediction pairs from physics, across the impact dimensions; 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 physics 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 February 2027. 11 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1O4 Mini73#372
2GPT-4.1 Mini70.6#799
3Qwen 2.5 14B68.4#1819
4Claude Haiku 4.568#587
5Qwen 2.5 7B66.1#2902
6Qwen 3 4B65.3#2379
7Llama 3.1 8B64.4#5193
8Ministral 3 3B61.9#3606
9Llama 3 8B61.6#5897
10Llama 3.2 3B54.1#7835
11nemotron-3-nano-30B-a3B52.5#1902

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=sciimpact-physics · How It Works · Data refreshed daily, snapshot 2026-11-22.