Llama 3 8B Instruct OAS: benchmark results
Provider: Meta.
Aggregate Rating 1419 ± 35, rank #5219 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - MMLU | 67.03 | Accuracy (%) (5-shot) | 88.7 |
| Open LLM Leaderboard v1 - GSM8K | 64.9 | Accuracy (%) (5-shot) | 79.1 |
| Open LLM Leaderboard v1 - ARC Challenge | 61.69 | Normalized accuracy (%) (25-shot) | 53.9 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 51.42 | MC2 (%) (0-shot) | 51.9 |
| Open LLM Leaderboard v1 - WinoGrande | 75.45 | Accuracy (%) (5-shot) | 42.2 |
| Open LLM Leaderboard v1 - HellaSwag | 78.45 | Normalized accuracy (%) (10-shot) | 33.9 |
Interactive version: theaggregate.ai/model?slug=llama-3-8b-instruct-oas · How It Works · Data refreshed daily, snapshot 2026-11-08.