NuclearQAv2 - Boolean: leaderboard

Metric: Accuracy (%) on 750 yes-or-no nuclear engineering questions (factual grounding); mean of three runs; answers requested without reasoning text. Source: arxiv.org. Saturation forecast: Estimated already saturated. 9 models tracked.

Top models

#ModelScore
1GPT-5.280.22
2GPT-5.479.91
3GPT-OSS-120B79.6
4NVIDIA-Nemotron-3-Super-120B-A12B-FP879.24
5Llama 3 70B Instruct76
6GPT-4o73.07
7Llama 3 8B Instruct71.73
8Nova Pro56.98
9Mistral 7B Instruct38.31

Interactive version: theaggregate.ai/benchmark?slug=nuclearqav2-boolean · How It Works · Data refreshed daily, snapshot 2026-09-29.