CodeLlama-7B-Instruct: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1204 ± 41, rank #9597 of 10916 rated models, from 29 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MERA Code - YABLoCo | 9.62 | pass@1 (%) | 74.6 |
| BigCode Models Leaderboard | 45.6 | HumanEval Python Pass@1 (%) | 54.2 |
| EvoEval Combine | 8 | Pass@1 (%) | 39 |
| EvoEval Verbose | 45.12 | Pass@1 (%) | 37 |
| EvoEval Tool Use | 44 | Pass@1 (%) | 32 |
| EvoEval | 31.73 | Pass@1 (%) | 30 |
| EvoEval Concise | 41.46 | Pass@1 (%) | 30 |
| EvoEval Creative | 18 | Pass@1 (%) | 27 |
| EvoEval Subtle | 40 | Pass@1 (%) | 27 |
| EvoEval Difficult | 14 | Pass@1 (%) | 26 |
| MERA Code - CodeLinterEval | 31.55 | pass@1 (%) | 24.6 |
| Big Code Memorization - HumanEval-ET pass@50 | 24.55 | HumanEval-ET pass@50 (%) | 23.5 |
Interactive version: theaggregate.ai/model?slug=codellama-7b-instruct · How It Works · Data refreshed daily, snapshot 2026-11-08.