llama2_tmt-13B-v1: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1394 ± 17, rank #1956 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-Winogrande | 67.01 | Accuracy (%) | 90.3 |
| Open Korean LLM - KorNAT-Harmless | 65.02 | Normalized accuracy (%) | 78.5 |
| Open Korean LLM - KorNAT-CKA | 24.77 | Normalized accuracy (%) | 53.4 |
| Open Korean LLM - Ko-EQ-Bench | 4.36 | EQ-Bench score | 50.5 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 12.59 | Exact match, flexible extract (%) | 39.4 |
| Open Korean LLM Leaderboard | 33.03 | Average Score (%) | 37.6 |
| Open Ko-LLM Leaderboard | 33.03 | Average (%) | 37.3 |
| Open Korean LLM - KorNAT-Helpful | 44.38 | Normalized accuracy (%) | 29.3 |
| Open Korean LLM - Ko-IFEval | 17.18 | Strict accuracy, prompt/instruction mean (%) | 12.6 |
| Open Korean LLM - Ko-GPQA-Diamond | 18.69 | Normalized accuracy (%) | 4.1 |
Interactive version: theaggregate.ai/model?slug=llama2-tmt-13b-v1 · How It Works · Data refreshed daily, snapshot 2026-09-19.