red-hat-ai/haiku-hard

Coding

Hard subset of SWE-style Go bug-fixing tasks drawn from Red Hat-ecosystem repos (openshift, operator-framework, coreos), filtered to instances Claude Haiku failed to solve.

← Back to Registry

Run this task

CLI:

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

Python:

from inspect_ai import eval
from inspect_harbor import red_hat_ai_haiku_hard

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

Dataset information

Harbor registry	red-hat-ai/haiku-hard
Inspect task	`red_hat_ai_haiku_hard`
Latest digest	sha256:ac5293f67b0debc77dd07d37c1f06a1cfaf42d857e34e158c6745045dfee5b52
Samples	138

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