llama3-deepspeed-v1.0: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1481 ± 17, rank #1275 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-IFEval | 44.49 | Strict accuracy, prompt/instruction mean (%) | 88.9 |
| Open Korean LLM - Ko-GPQA-Diamond | 25.76 | Normalized accuracy (%) | 81.3 |
| Open Korean LLM Leaderboard | 40.48 | Average Score (%) | 77.3 |
| Open Ko-LLM Leaderboard | 40.48 | Average (%) | 77.2 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 39.73 | Exact match, flexible extract (%) | 74 |
| Open Korean LLM - KorNAT-Helpful | 48.15 | Normalized accuracy (%) | 65.1 |
| Open Korean LLM - Ko-EQ-Bench | 16.57 | EQ-Bench score | 61.9 |
| Open Korean LLM - KorNAT-CKA | 25.38 | Normalized accuracy (%) | 57.6 |
| Open Korean LLM - Ko-Winogrande | 60.06 | Accuracy (%) | 51.1 |
| Open Korean LLM - KorNAT-Harmless | 62.01 | Normalized accuracy (%) | 36.9 |
Interactive version: theaggregate.ai/model?slug=llama3-deepspeed-v1-0 · How It Works · Data refreshed daily, snapshot 2026-09-19.