scale-ai/swe-atlas-rf
Coding
SWE-Atlas - Refactoring – A benchmark of refactoring tasks for coding agents.
Run this task
CLI:
inspect eval inspect_harbor/scale_ai_swe_atlas_rf --model openai/gpt-5Python:
from inspect_ai import eval
from inspect_harbor import scale_ai_swe_atlas_rf
eval(scale_ai_swe_atlas_rf(), model="openai/gpt-5")Dataset information
| Harbor registry | scale-ai/swe-atlas-rf |
| Inspect task | scale_ai_swe_atlas_rf |
| Latest digest | sha256:5f65e33ebd80c3998b2f82d6c78349da1acce1823e4b18e4851f980a03b2ad93 |
| Samples | 70 |
| Source | https://github.com/scaleapi/SWE-Atlas |
See Task Parameters for the parameter set shared across all Harbor tasks.