Skip to main content

Attributes

UUID
Unique identifier (UUID) of the dataset item.
dict[str, Any]
Input data for the test case; passed to the evaluation task function.
dict[str, Any] | None
Expected (ground-truth) outputs for the test case, used by evaluators for comparison.
dict
Additional descriptive test case metadata, e.g. {"difficulty": "easy"}; also passed to the evaluation task function.
dict
Any other custom data for the test case; also passed to the evaluation task function.
str | None
Optional name identifying where the test case came from.
str | None
Optional identifier of the test case within its source.
datetime
Timestamp when the item was created on the server.
datetime
Timestamp when the item was last updated on the server.