PRL-Bench - Statistical Physics: leaderboard
PRL-Bench (2026) builds long-horizon theoretical and computational physics research tasks from 100 Physical Review Letters papers; a GPT-5 judge awards pre-set points for final answers and intermediate rubric checkpoints, mean of five runs, with a code interpreter and no search. This board covers the statistical physics and complex systems tasks, on a 0-100 scale; higher is better.
Metric: Score (0-100) on the statistical physics and complex systems (Stat) tasks, research tasks built from 100 Physical Review Letters papers; a GPT-5 judge awards pre-assigned points for final answers and intermediate rubric checkpoints, summed and normalized to 0-100, mean of five runs; code interpreter allowed, search disabled; higher is better. Source: arxiv.org. Saturation forecast: Around April 2027. 6 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 40.76 | #99 |
| 2 | GPT-5.4 | 33.88 | #135 |
| 3 | Claude Opus 4.6 | 32.1 | #113 |
| 4 | Qwen 3.5 Plus | 25.87 | #242 |
| 5 | Kimi K2.5 | 25.71 | #266 |
| 6 | Seed 2.0 Pro | 24.94 | #222 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=prl-bench-statistical-physics · How It Works · Data refreshed daily, snapshot 2026-11-21.