Llama 2 3B Base: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1093 ± 39, rank #10780 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 50.71 | MC2 (%) (0-shot) | 49.3 |
| Open LLM Leaderboard v1 - ARC Challenge | 26.96 | Normalized accuracy (%) (25-shot) | 6.1 |
| Open LLM Leaderboard v1 - GSM8K | 0 | Accuracy (%) (5-shot) | 4.5 |
| Open LLM Leaderboard v1 - HellaSwag | 26.52 | Normalized accuracy (%) (10-shot) | 3.3 |
| Open LLM Leaderboard v1 - WinoGrande | 49.64 | Accuracy (%) (5-shot) | 2.2 |
| Open LLM Leaderboard v1 - MMLU | 23.33 | Accuracy (%) (5-shot) | 1.9 |
Interactive version: theaggregate.ai/model?slug=llama-2-3b-base · How It Works · Data refreshed daily, snapshot 2026-11-08.