scale-ai/swe-bench-pro
Coding
SWE-Bench-Pro: long-horizon enterprise software engineering tasks.
Run this task
CLI:
inspect eval inspect_harbor/scale_ai_swe_bench_pro --model openai/gpt-5Python:
from inspect_ai import eval
from inspect_harbor import scale_ai_swe_bench_pro
eval(scale_ai_swe_bench_pro(), model="openai/gpt-5")Dataset information
| Harbor registry | scale-ai/swe-bench-pro |
| Inspect task | scale_ai_swe_bench_pro |
| Latest digest | sha256:88411d32ff27e53a4c1a7e29f0c2aeba180c8e5d60f221cab5ed56325f33549d |
| Samples | 731 |
| Paper | arxiv |
| Source | https://github.com/scaleapi/SWE-bench_Pro-os |
See Task Parameters for the parameter set shared across all Harbor tasks.