llama3_Commonsense (Reasoning): benchmark results
Provider: Meta. Access: Open.
Unified ELO 1494 ± 1, rank #1524 of 3078 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-CKA | 32.84 | Normalized accuracy (%) | 89.8 |
| Open Korean LLM - Ko-GPQA-Diamond | 26.77 | Normalized accuracy (%) | 88.9 |
| Open Korean LLM - KorNAT-Harmless | 64.95 | Normalized accuracy (%) | 78.1 |
| Open Korean LLM - Ko-IFEval | 36.68 | Strict accuracy, prompt/instruction mean (%) | 76.3 |
| Open Korean LLM - KorNAT-Helpful | 47.6 | Normalized accuracy (%) | 57.8 |
| Open Korean LLM - Ko-EQ-Bench | 3.6 | EQ-Bench score | 49.7 |
| Open Korean LLM Leaderboard | 35.09 | Average Score (%) | 49.6 |
| Open Ko-LLM Leaderboard | 35.09 | Average (%) | 49.4 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 13.34 | Exact match, flexible extract (%) | 40.8 |
| Open Korean LLM - Ko-Winogrande | 56.59 | Accuracy (%) | 27.1 |
Interactive version: theaggregate.ai/model?slug=llama3-commonsense-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-19.