NuclearQAv2 - Verbal: leaderboard
Metric: Accuracy (%) on 283 short-answer concept questions, judged semantically equivalent to the reference by an LLM judge; mean of three runs; answers requested without reasoning text. Source: arxiv.org. Saturation forecast: Around December 2026. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.2 | 82.33 |
| 2 | GPT-5.4 | 81.04 |
| 3 | NVIDIA-Nemotron-3-Super-120B-A12B-FP8 | 79.27 |
| 4 | GPT-OSS-120B | 78.33 |
| 5 | GPT-4o | 74.79 |
| 6 | Nova Pro | 70.79 |
| 7 | Llama 3 70B Instruct | 67.49 |
| 8 | Mistral 7B Instruct | 51.59 |
| 9 | Llama 3 8B Instruct | 48.65 |
Interactive version: theaggregate.ai/benchmark?slug=nuclearqav2-verbal · How It Works · Data refreshed daily, snapshot 2026-09-29.