llama-ko-7B Benchmark Results

llama-ko-7B: benchmark results

Provider: Meta. Access: Open.

Unified ELO 1301 ± 17, rank #2455 of 2656 rated models, from 10 benchmark results.

Benchmark	Score	Metric	Percentile
Open Korean LLM - Ko-GPQA-Diamond	27.27	Normalized accuracy (%)	92.4
Open Korean LLM - KorNAT-Harmless	61.99	Normalized accuracy (%)	36.9
Open Korean LLM - KorNAT-Helpful	44.47	Normalized accuracy (%)	29.9
Open Korean LLM - Ko-EQ-Bench	0	EQ-Bench score	24.5
Open Korean LLM - Ko-IFEval	16.94	Strict accuracy, prompt/instruction mean (%)	12.3
Open Korean LLM Leaderboard	28.53	Average Score (%)	6.8
Open Ko-LLM Leaderboard	28.53	Average (%)	6.6
Open Korean LLM - Ko-GSM8K (flexible extract)	0	Exact match, flexible extract (%)	2
Open Korean LLM - Ko-Winogrande	49.72	Accuracy (%)	2
Open Korean LLM - KorNAT-CKA	5.39	Normalized accuracy (%)	0

Interactive version: theaggregate.ai/model?slug=llama-ko-7b · How It Works · Data refreshed daily, snapshot 2026-09-19.