Llama-3-Refueled: benchmark results
Provider: Meta. Released 2024-05-08. Access: Open.
Aggregate Rating 1407 ± 44, rank #5764 of 10916 rated models, from 14 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - MuSR | 44.54 | Score | 81.2 |
| Open LLM Leaderboard - BBH | 58.71 | Score | 79.9 |
| Open Chinese LLM - C-Eval Semantic | 71.28 | Accuracy (%) | 56.2 |
| Open LLM Leaderboard - GPQA | 29.95 | Score | 55.6 |
| Open LLM Leaderboard - IFEval | 46.2 | Score | 52.1 |
| Open LLM Leaderboard - MMLU-Pro | 30.95 | Score | 39.8 |
| Open Chinese LLM - CMMLU | 51.67 | Accuracy (%) | 39.3 |
| Open LLM Leaderboard - MATH Level 5 | 6.65 | Score | 35.9 |
| Open Chinese LLM - GSM8K | 37.45 | Accuracy (%) | 34.6 |
| Open Chinese LLM Leaderboard | 51.22 | Official Nonzero-Task Mean (%) | 21.3 |
| Open Chinese LLM - WinoGrande | 60.62 | Accuracy (%) | 18.5 |
| Open Chinese LLM - HellaSwag | 51.62 | Accuracy (%) | 9.8 |
Interactive version: theaggregate.ai/model?slug=llama-3-refueled · How It Works · Data refreshed daily, snapshot 2026-11-08.