ade-bench@1.0
Coding
Analytics Data Engineer Bench: tasks evaluating AI agents on dbt/SQL data analytics engineering bugs.
Run this task
CLI:
inspect eval inspect_harbor/ade_bench_1_0 --model openai/gpt-5Python:
from inspect_ai import eval
from inspect_harbor import ade_bench_1_0
eval(ade_bench_1_0(), model="openai/gpt-5")Dataset information
| Harbor registry | ade-bench@1.0 |
| Inspect task | ade_bench_1_0 |
| Version | 1.0 |
| Samples | 48 |
| Source | https://github.com/dbt-labs/ade-bench |
See Task Parameters for the parameter set shared across all Harbor tasks.