Llama 3 8B Ko Sft V1: benchmark results

Provider: Meta.

Aggregate Rating 1440 ± 43, rank #4374 of 10556 rated models, from 9 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open Korean LLM Leaderboard41.71Average Score (%)83.4
Open Korean LLM - Ko-GSM8K (flexible extract)48.98Exact match, flexible extract (%)82.5
Open Korean LLM - Ko-IFEval37.26Strict accuracy, prompt/instruction mean (%)77.7
Open Korean LLM - Ko-EQ-Bench25.96EQ-Bench score72.3
Open Korean LLM - KorNAT-CKA24.8Normalized accuracy (%)53.6
Open Korean LLM - Ko-Winogrande60.14Accuracy (%)51.6
Open Korean LLM - KorNAT-Harmless62.91Normalized accuracy (%)51.6
Open Korean LLM - KorNAT-Helpful46.33Normalized accuracy (%)44.8
Open Korean LLM - Ko-GPQA-Diamond21.21Normalized accuracy (%)24

Interactive version: theaggregate.ai/model?slug=llama-3-8b-ko-sft-v1 · How It Works · Data refreshed daily, snapshot 2026-11-17.