llama3-diverce-v4.5: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1481 ± 17, rank #1276 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-IFEval | 44.24 | Strict accuracy, prompt/instruction mean (%) | 88.6 |
| Open Korean LLM Leaderboard | 40.68 | Average Score (%) | 78.6 |
| Open Ko-LLM Leaderboard | 40.68 | Average (%) | 78.3 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 42.61 | Exact match, flexible extract (%) | 77.6 |
| Open Korean LLM - Ko-EQ-Bench | 17.52 | EQ-Bench score | 63.3 |
| Open Korean LLM - Ko-Winogrande | 60.06 | Accuracy (%) | 51.1 |
| Open Korean LLM - Ko-GPQA-Diamond | 23.23 | Normalized accuracy (%) | 48.1 |
| Open Korean LLM - KorNAT-CKA | 23.55 | Normalized accuracy (%) | 45.8 |
| Open Korean LLM - KorNAT-Helpful | 46.38 | Normalized accuracy (%) | 45.4 |
| Open Korean LLM - KorNAT-Harmless | 60.99 | Normalized accuracy (%) | 21.8 |
Interactive version: theaggregate.ai/model?slug=llama3-diverce-v4-5 · How It Works · Data refreshed daily, snapshot 2026-09-19.