Logical-CommonsenseQA (3-Shot) - AND: leaderboard

Metric: Macro-F1 (%; both composed statements must be plausible; three labelled examples in the prompt; 250 human-validated four-option test questions). Source: arxiv.org. Saturation forecast: Estimated already saturated. 3 models tracked.

Top models

#ModelScore
1Llama 3.3 70B Instruct85.7
2Qwen 2.5 7B Instruct77.9
3Llama 3.1 8B Instruct71.3

Interactive version: theaggregate.ai/benchmark?slug=logical-commonsenseqa-3-shot-and · How It Works · Data refreshed daily, snapshot 2026-09-26.