NSMQ Riddles - Chemistry: leaderboard

Metric: Exact match (%; 44 chemistry riddles from the 2019 NSMQ riddles round, all clues given at once, answer checked against the answer and its listed variants). Source: arxiv.org. Saturation forecast: Around March 2028. 12 models tracked.

Top models

#ModelScore
1GPT-5.4 (High)75
2Claude Opus 4.6 (High)72.73
3GPT-5.4 (Low)72.73
4GPT-OSS-120B (High)68.18
5Gemini 3.1 Pro (Preview) (High)68.18
6GPT-OSS-120B (Low)65.91
7Gemini 3.1 Pro (Preview) (Low)65.91

Interactive version: theaggregate.ai/benchmark?slug=nsmq-riddles-chemistry · How It Works · Data refreshed daily, snapshot 2026-09-25.