Llama 3 8B Instruct 1048k: benchmark results
Provider: Meta. Released 2024-04-18.
Aggregate Rating 1375 ± 35, rank #7008 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 44.35 | Accuracy (%) (5-shot) | 58.4 |
| Open LLM Leaderboard v1 - MMLU | 61.92 | Accuracy (%) (5-shot) | 53.3 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 49.26 | MC2 (%) (0-shot) | 44.2 |
| Open LLM Leaderboard v1 - ARC Challenge | 54.44 | Normalized accuracy (%) (25-shot) | 31.4 |
| Open LLM Leaderboard v1 - HellaSwag | 76.77 | Normalized accuracy (%) (10-shot) | 29.5 |
| Open LLM Leaderboard v1 - WinoGrande | 72.3 | Accuracy (%) (5-shot) | 28.8 |
Interactive version: theaggregate.ai/model?slug=llama-3-8b-instruct-1048k · How It Works · Data refreshed daily, snapshot 2026-11-08.