PhysDox (Self-Consistency): leaderboard

Metric: Stage-1 severity-detection macro-F1 (0-100) on the 683-protocol expert-curated Gold set of biomedical physiological-sensing protocols (six domains): classify each protocol as valid, minor issue or fatal physical-feasibility violation; self-consistency: majority vote over three temperature-0.7 samples; higher is better. Source: arxiv.org. Saturation forecast: Around 2031. 6 models tracked.

Top models

#ModelScore
1DeepSeek V4 Flash53
2Claude Opus 4.751.4
3MiMo-V2.546.3
4DeepSeek V4 Pro45.5
5MiMo-V2.5-Pro44.7
6GPT-5.538.5

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