Llama 3 8B Magpie Align SFT V0.3: benchmark results
Provider: Meta. Released 2024-07-13. Access: Open.
Unified ELO 1445 ± 1, rank #967 of 1392 rated models, from 15 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM Leaderboard | 42.75 | Average Score (%) | 84.6 |
| Open LLM Leaderboard - IFEval | 50.64 | Score | 60.2 |
| Open Chinese LLM - CMMLU | 53.47 | Accuracy (%) | 50.6 |
| Open Chinese LLM - C-Eval Semantic | 68.59 | Accuracy (%) | 46.4 |
| Open Chinese LLM - WinoGrande | 62.59 | Accuracy (%) | 45.8 |
| Open Chinese LLM - GSM8K | 42.99 | Accuracy (%) | 43.9 |
| Open LLM Leaderboard - MATH Level 5 | 7.33 | Score | 38.6 |
| Open LLM Leaderboard - BBH | 23.7 | Score | 35.8 |
| Open Chinese LLM Leaderboard | 54.83 | Average Score (%) | 35.6 |
| Open LLM Leaderboard - MMLU-Pro | 21.14 | Score | 33.5 |
| Open Chinese LLM - ARC Challenge | 47.78 | Accuracy (%) | 31.6 |
| Open Chinese LLM - HellaSwag | 56.88 | Accuracy (%) | 30.6 |
Interactive version: theaggregate.ai/model?slug=llama-3-8b-magpie-align-sft-v0-3 · How It Works · Data refreshed daily, snapshot 2026-09-05.