CodeLlama-34B-Python: benchmark results
Provider: Meta.
Aggregate Rating 1362 ± 51, rank #7373 of 10916 rated models, from 11 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BigCode Models Leaderboard | 53.3 | HumanEval Python Pass@1 (%) | 69.5 |
| ENAMEL | 26.8 | Greedy eff@1 (%) (imported) | 65.5 |
| CRUXEval | 41.4 | Output Prediction pass@1 (%) | 50 |
| EvoEval Combine | 11 | Pass@1 (%) | 50 |
| EvoEval Subtle | 47 | Pass@1 (%) | 48 |
| EvoEval Concise | 49.39 | Pass@1 (%) | 46 |
| EvoEval Creative | 24 | Pass@1 (%) | 46 |
| EvoEval | 35.61 | Pass@1 (%) | 40 |
| EvoEval Tool Use | 44 | Pass@1 (%) | 32 |
| EvoEval Verbose | 42.07 | Pass@1 (%) | 32 |
| EvoEval Difficult | 15 | Pass@1 (%) | 28 |
Interactive version: theaggregate.ai/model?slug=codellama-34b-python · How It Works · Data refreshed daily, snapshot 2026-11-08.