Llama 3 8B Instruct 262k [bfloat16] [rev: 9834c466d6ed]: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1403 ± 34, rank #5958 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 61.79 | Accuracy (%) (5-shot) | 74.6 |
| Open LLM Leaderboard v1 - MMLU | 64.27 | Accuracy (%) (5-shot) | 67.4 |
| Open LLM Leaderboard v1 - WinoGrande | 75.77 | Accuracy (%) (5-shot) | 44 |
| Open LLM Leaderboard v1 - ARC Challenge | 59.3 | Normalized accuracy (%) (25-shot) | 42.8 |
| Open LLM Leaderboard v1 - HellaSwag | 80.84 | Normalized accuracy (%) (10-shot) | 42.6 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 44.32 | MC2 (%) (0-shot) | 26.7 |
Interactive version: theaggregate.ai/model?slug=llama-3-8b-instruct-262k-bfloat16-rev-9834c466d6ed · How It Works · Data refreshed daily, snapshot 2026-11-08.