KnotBench - Same Knot (PD Codes): leaderboard
Metric: Accuracy (%) on the 200 A0-S items: whether two planar-diagram (PD) codes describe the same knot; zero-shot with a forced ANSWER line and strict matching, 64K output-token budget, random baseline 50%; higher is better. Source: arxiv.org. Saturation forecast: Around April 2027. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.7 (Thinking) | 60 |
| 2 | GPT-5 (Thinking) | 55.5 |
| 3 | Claude Opus 4.7 | 52 |
Interactive version: theaggregate.ai/benchmark?slug=knotbench-same-knot-pd-codes · How It Works · Data refreshed daily, snapshot 2026-10-07.