featurebench-lite-modal@1.0

Coding

FeatureBench lite split for Modal: 30 feature-implementation tasks with gpus=1 for GPU tasks (7/30). Use with -e modal.

← Back to Registry

Run this task

CLI:

inspect eval inspect_harbor/featurebench_lite_modal_1_0 --model openai/gpt-5

Python:

from inspect_ai import eval
from inspect_harbor import featurebench_lite_modal_1_0

eval(featurebench_lite_modal_1_0(), model="openai/gpt-5")

Dataset information

Harbor registry featurebench-lite-modal@1.0
Inspect task featurebench_lite_modal_1_0
Version 1.0
Samples 30
Paper arxiv

See Task Parameters for the parameter set shared across all Harbor tasks.