KnotBench - Reidemeister Move (PD Codes): leaderboard

Metric: Accuracy (%) on the 100 B0-S items: which Reidemeister move connects two consecutive PD codes; zero-shot with a forced ANSWER line and strict matching, 64K output-token budget, random baseline 16.7%; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 4 models tracked.

Top models

#ModelScore
1GPT-5 (Thinking)88
2Claude Opus 4.784
3Claude Opus 4.7 (Thinking)84

Interactive version: theaggregate.ai/benchmark?slug=knotbench-reidemeister-move-pd-codes · How It Works · Data refreshed daily, snapshot 2026-10-07.