CodeLlama-70B-hf: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1361 ± 33, rank #7403 of 10916 rated models, from 9 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EffiBench - NMU | 1.9 | Normalized Memory Usage | 79.3 |
| Open LLM Leaderboard v1 - GSM8K | 43.97 | Accuracy (%) (5-shot) | 58.1 |
| Open LLM Leaderboard v1 - MMLU | 59.67 | Accuracy (%) (5-shot) | 45.6 |
| Open LLM Leaderboard v1 - WinoGrande | 75.22 | Accuracy (%) (5-shot) | 40.9 |
| Open LLM Leaderboard v1 - ARC Challenge | 56.74 | Normalized accuracy (%) (25-shot) | 36 |
| Open LLM Leaderboard v1 - HellaSwag | 78.21 | Normalized accuracy (%) (10-shot) | 33.2 |
| EffiBench - NET | 3.19 | Normalized Execution Time | 18.3 |
| BiGGen-Bench | 2.59 | Average Score (1-5) | 16.7 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 39.79 | MC2 (%) (0-shot) | 12 |
Interactive version: theaggregate.ai/model?slug=codellama-70b-hf · How It Works · Data refreshed daily, snapshot 2026-11-08.