Llama 2 4B: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1124 ± 35, rank #10457 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 0.45 | Accuracy (%) (5-shot) | 12.3 |
| Open LLM Leaderboard v1 - HellaSwag | 53.29 | Normalized accuracy (%) (10-shot) | 11.3 |
| Open LLM Leaderboard v1 - WinoGrande | 57.46 | Accuracy (%) (5-shot) | 10.7 |
| Open LLM Leaderboard v1 - ARC Challenge | 31.23 | Normalized accuracy (%) (25-shot) | 10.1 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 38.72 | MC2 (%) (0-shot) | 8.8 |
| Open LLM Leaderboard v1 - MMLU | 24.22 | Accuracy (%) (5-shot) | 3.3 |
Interactive version: theaggregate.ai/model?slug=llama-2-4b · How It Works · Data refreshed daily, snapshot 2026-11-08.