Llama 3 8B DUS Initialized: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1325 ± 35, rank #8253 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - WinoGrande | 75.93 | Accuracy (%) (5-shot) | 44.9 |
| Open LLM Leaderboard v1 - MMLU | 53.68 | Accuracy (%) (5-shot) | 33.1 |
| Open LLM Leaderboard v1 - GSM8K | 9.33 | Accuracy (%) (5-shot) | 29.9 |
| Open LLM Leaderboard v1 - HellaSwag | 75.78 | Normalized accuracy (%) (10-shot) | 27.3 |
| Open LLM Leaderboard v1 - ARC Challenge | 49.57 | Normalized accuracy (%) (25-shot) | 24.1 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 41.72 | MC2 (%) (0-shot) | 17.9 |
Interactive version: theaggregate.ai/model?slug=llama-3-8b-dus-initialized · How It Works · Data refreshed daily, snapshot 2026-11-08.