Llama 2 13B Test V0.1: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1383 ± 17, rank #2031 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-Winogrande | 64.33 | Accuracy (%) | 78 |
| Open Korean LLM - Ko-IFEval | 27.57 | Strict accuracy, prompt/instruction mean (%) | 44 |
| Open Korean LLM - KorNAT-Harmless | 62.44 | Normalized accuracy (%) | 44 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 12.36 | Exact match, flexible extract (%) | 38.9 |
| Open Korean LLM - KorNAT-CKA | 21.29 | Normalized accuracy (%) | 33.8 |
| Open Korean LLM Leaderboard | 32.27 | Average Score (%) | 32.3 |
| Open Ko-LLM Leaderboard | 32.27 | Average (%) | 32.1 |
| Open Korean LLM - KorNAT-Helpful | 44.65 | Normalized accuracy (%) | 31.6 |
| Open Korean LLM - Ko-GPQA-Diamond | 21.21 | Normalized accuracy (%) | 24 |
| Open Korean LLM - Ko-EQ-Bench | -4.05 | EQ-Bench score | 2.6 |
Interactive version: theaggregate.ai/model?slug=llama-2-13b-test-v0-1 · How It Works · Data refreshed daily, snapshot 2026-09-19.