Llama 3 8B Slerp 262k Sft Lora Ko V2: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1467 ± 17, rank #1400 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GSM8K (flexible extract) | 41.17 | Exact match, flexible extract (%) | 75.7 |
| Open Korean LLM Leaderboard | 39.5 | Average Score (%) | 72.5 |
| Open Ko-LLM Leaderboard | 39.5 | Average (%) | 72.2 |
| Open Korean LLM - Ko-EQ-Bench | 17.77 | EQ-Bench score | 63.7 |
| Open Korean LLM - Ko-IFEval | 32.25 | Strict accuracy, prompt/instruction mean (%) | 61.8 |
| Open Korean LLM - Ko-Winogrande | 61.17 | Accuracy (%) | 60.3 |
| Open Korean LLM - KorNAT-Helpful | 47.15 | Normalized accuracy (%) | 53.5 |
| Open Korean LLM - KorNAT-Harmless | 62.93 | Normalized accuracy (%) | 52.2 |
| Open Korean LLM - KorNAT-CKA | 21.55 | Normalized accuracy (%) | 35.4 |
| Open Korean LLM - Ko-GPQA-Diamond | 22.22 | Normalized accuracy (%) | 35 |
Interactive version: theaggregate.ai/model?slug=llama-3-8b-slerp-262k-sft-lora-ko-v2 · How It Works · Data refreshed daily, snapshot 2026-09-19.