Function completion — Given a Python function signature and docstring, write the body that passes the tests
- Category: Coding
- Metric: pass@1
- Source benchmark: HumanEval+
- Examples: 10
items.jsonl— 10 items, one JSON object per line (input+expected_output).
Try it live in the Pareta app → pick this task → Try the example set, or download items.jsonl and upload it to your own eval set.