Epoch AI - Common Sense Qa 2: leaderboard
Commonsense reasoning through multiple-choice questions about objects, social situations, and cause-and-effect, with carefully constructed distractors that resist superficial shortcuts.
Metric: Score. Source: epoch.ai. Saturation forecast: Estimated already saturated. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Median Human | 94.1 | |
| 2 | T5 (11B) | 67.8 | #1652 |
| 3 | GPT-3.5 Turbo (0613) | 57 | #1118 |
| 4 | text-davinci-001 | 52.9 | #1512 |
| 5 | Llama 2 70B Base | 50 | #1025 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=epoch-ai-common-sense-qa-2 · How It Works · Data refreshed daily, snapshot 2026-10-20.