Llama 2 13B Math1.2 [bfloat16] [rev: B05b4c22893e]: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1340 ± 34, rank #7978 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - HellaSwag | 80.61 | Normalized accuracy (%) (10-shot) | 41.6 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 48.3 | MC2 (%) (0-shot) | 40.6 |
| Open LLM Leaderboard v1 - ARC Challenge | 57.08 | Normalized accuracy (%) (25-shot) | 36.6 |
| Open LLM Leaderboard v1 - WinoGrande | 74.27 | Accuracy (%) (5-shot) | 35.8 |
| Open LLM Leaderboard v1 - MMLU | 53.05 | Accuracy (%) (5-shot) | 32.3 |
| Open LLM Leaderboard v1 - GSM8K | 10.99 | Accuracy (%) (5-shot) | 32.2 |
Interactive version: theaggregate.ai/model?slug=llama-2-13b-math1-2-bfloat16-rev-b05b4c22893e · How It Works · Data refreshed daily, snapshot 2026-11-08.