abundant/swe-marathon

Coding

Ultra long-horizon software engineering tasks from SWE-Marathon.

← Back to Registry

Run this task

CLI:

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

Python:

from inspect_ai import eval
from inspect_harbor import abundant_swe_marathon

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

Dataset information

Harbor registry abundant/swe-marathon
Inspect task abundant_swe_marathon
Latest digest sha256:9b351bf9bd305e6fd8bfdba9120e42551d027bf173b81ea790e8df2f21d7a624
Samples 20
Paper arxiv
Source https://github.com/abundant-ai/swe-marathon

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