CodeLlama-13B-Python-fp16: benchmark results
TheBloke's Transformers-format FP16 conversion of Meta's CodeLlama-13B-Python, converted from Meta's original weights. The Instruct and Python variants remain separate releases. Provider: TheBloke. Access: Open.
Aggregate Rating 1145 ± 17, rank #10039 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - Moral Scenarios | 27.15 | Accuracy (%) | 58.3 |
| MMLU-by-task - High School Statistics | 39.81 | Accuracy (%) | 54.3 |
| MMLU-by-task - High School Physics | 29.8 | Accuracy (%) | 46.5 |
| MMLU-by-task - TruthfulQA MC2 | 43.99 | Accuracy (%) | 46.4 |
| MMLU-by-task - TruthfulQA MC1 | 26.56 | Accuracy (%) | 41.9 |
| MMLU-by-task - High School Mathematics | 26.3 | Accuracy (%) | 40 |
| MMLU-by-task - Econometrics | 27.19 | Accuracy (%) | 39.6 |
| MMLU-by-task - College Mathematics | 30 | Accuracy (%) | 38.1 |
| MMLU-by-task - Security Studies | 38.78 | Accuracy (%) | 36.6 |
| MMLU-by-task - Professional Accounting | 28.37 | Accuracy (%) | 34.2 |
| MMLU-by-task - Miscellaneous | 31.8 | Accuracy (%) | 32.2 |
| MMLU-by-task - Anatomy | 33.33 | Accuracy (%) | 32.1 |
Interactive version: theaggregate.ai/model?slug=codellama-13b-python-fp16 · How It Works · Data refreshed daily, snapshot 2026-11-08.