Llama 3 8B Instruct Gradient 4194k: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1367 ± 35, rank #7248 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 46.1 | Accuracy (%) (5-shot) | 59.9 |
| Open LLM Leaderboard v1 - MMLU | 60.1 | Accuracy (%) (5-shot) | 46.9 |
| Open LLM Leaderboard v1 - HellaSwag | 77.89 | Normalized accuracy (%) (10-shot) | 32.2 |
| Open LLM Leaderboard v1 - ARC Challenge | 53.75 | Normalized accuracy (%) (25-shot) | 30.1 |
| Open LLM Leaderboard v1 - WinoGrande | 72.53 | Accuracy (%) (5-shot) | 29.3 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 44.82 | MC2 (%) (0-shot) | 28.7 |
Interactive version: theaggregate.ai/model?slug=llama-3-8b-instruct-gradient-4194k · How It Works · Data refreshed daily, snapshot 2026-11-08.