KnotBench - Same Crossing Count (Images): leaderboard

The A2-I task of KnotBench (2026, The Gordian Knot for VLMs): judge whether two diagrams of one knot and chirality have the same number of crossings, with answers checked against Regina canonical knot signatures. The score is accuracy in percent under strict answer matching, with a random baseline of 50 percent; higher is better.

Metric: Accuracy (%) on the 100 A2-I items: whether two diagrams of one knot and chirality have the same number of crossings; 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 February 2027. 2 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1Claude Opus 4.767#85
2GPT-5Thinking50#178

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=knotbench-same-crossing-count-images · How It Works · Data refreshed daily, snapshot 2026-11-21.