Llama 2 7B 32K Instruct [8bit] [rev: B050a6f17d46]: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1293 ± 35, rank #8835 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - HellaSwag | 78.47 | Normalized accuracy (%) (10-shot) | 34 |
| Open LLM Leaderboard v1 - WinoGrande | 72.85 | Accuracy (%) (5-shot) | 30.3 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 45.01 | MC2 (%) (0-shot) | 29.4 |
| Open LLM Leaderboard v1 - ARC Challenge | 51.37 | Normalized accuracy (%) (25-shot) | 25.7 |
| Open LLM Leaderboard v1 - MMLU | 45.53 | Accuracy (%) (5-shot) | 24 |
| Open LLM Leaderboard v1 - GSM8K | 4.7 | Accuracy (%) (5-shot) | 23.3 |
Interactive version: theaggregate.ai/model?slug=llama-2-7b-32k-instruct-8bit-rev-b050a6f17d46 · How It Works · Data refreshed daily, snapshot 2026-11-08.