Llama 3 13B: benchmark results
Provider: Meta.
Aggregate Rating 1364 ± 38, rank #7334 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - MMLU | 62.12 | Accuracy (%) (5-shot) | 54.1 |
| Open LLM Leaderboard v1 - GSM8K | 21.91 | Accuracy (%) (5-shot) | 42.2 |
| Open LLM Leaderboard v1 - HellaSwag | 80.66 | Normalized accuracy (%) (10-shot) | 41.8 |
| Open LLM Leaderboard v1 - ARC Challenge | 52.99 | Normalized accuracy (%) (25-shot) | 28.3 |
| Open LLM Leaderboard v1 - WinoGrande | 70.72 | Accuracy (%) (5-shot) | 25.2 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 39.28 | MC2 (%) (0-shot) | 10.6 |
Interactive version: theaggregate.ai/model?slug=llama-3-13b · How It Works · Data refreshed daily, snapshot 2026-11-08.