Llama 3 10B IT Ko 2024 0527: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1414 ± 26, rank #5457 of 10916 rated models, from 15 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-EQ-Bench | 31.15 | EQ-Bench score | 80.3 |
| Open Korean LLM Leaderboard | 40.67 | Average Score (%) | 78.4 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 42.46 | Exact match, flexible extract (%) | 77.3 |
| Open LLM Leaderboard v1 - MMLU | 64.49 | Accuracy (%) (5-shot) | 71.3 |
| Open Korean LLM - Ko-Winogrande | 62.27 | Accuracy (%) | 67.7 |
| Open LLM Leaderboard v1 - GSM8K | 53.83 | Accuracy (%) (5-shot) | 65.6 |
| Open Korean LLM - KorNAT-CKA | 25.45 | Normalized accuracy (%) | 58.3 |
| Open Korean LLM - Ko-GPQA-Diamond | 23.74 | Normalized accuracy (%) | 55.6 |
| Open LLM Leaderboard v1 - WinoGrande | 77.19 | Accuracy (%) (5-shot) | 53.9 |
| Open Korean LLM - Ko-IFEval | 29.75 | Strict accuracy, prompt/instruction mean (%) | 52.8 |
| Open LLM Leaderboard v1 - ARC Challenge | 59.47 | Normalized accuracy (%) (25-shot) | 43.5 |
| Open LLM Leaderboard v1 - HellaSwag | 79.69 | Normalized accuracy (%) (10-shot) | 38.8 |
Interactive version: theaggregate.ai/model?slug=llama-3-10b-it-ko-2024-0527 · How It Works · Data refreshed daily, snapshot 2026-11-08.