Llama 3 8B IT Kor Extented Chang: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1429 ± 25, rank #4698 of 10916 rated models, from 15 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GPQA-Diamond | 26.77 | Normalized accuracy (%) | 88.9 |
| Open LLM Leaderboard v1 - MMLU | 66.61 | Accuracy (%) (5-shot) | 87.7 |
| Open Korean LLM - Ko-EQ-Bench | 34.27 | EQ-Bench score | 84.1 |
| Open Korean LLM Leaderboard | 41.03 | Average Score (%) | 80.2 |
| Open Korean LLM - Ko-IFEval | 37.56 | Strict accuracy, prompt/instruction mean (%) | 78.7 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 41.02 | Exact match, flexible extract (%) | 75.4 |
| Open LLM Leaderboard v1 - GSM8K | 59.74 | Accuracy (%) (5-shot) | 71.6 |
| Open LLM Leaderboard v1 - WinoGrande | 77.58 | Accuracy (%) (5-shot) | 57 |
| Open Korean LLM - KorNAT-CKA | 25.18 | Normalized accuracy (%) | 55.6 |
| Open Korean LLM - Ko-Winogrande | 60.54 | Accuracy (%) | 55.1 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 51.56 | MC2 (%) (0-shot) | 52.4 |
| Open LLM Leaderboard v1 - ARC Challenge | 60.84 | Normalized accuracy (%) (25-shot) | 49.8 |
Interactive version: theaggregate.ai/model?slug=llama-3-8b-it-kor-extented-chang · How It Works · Data refreshed daily, snapshot 2026-11-08.