scale-ai/swe-atlas-qna

Coding

SWE-Atlas - Codebase QnA is a benchmark of deep codebase comprehension and QnA problems for coding agents. Checkout for instructions on running it.

← Back to Registry

Run this task

CLI:

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

Python:

from inspect_ai import eval
from inspect_harbor import scale_ai_swe_atlas_qna

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

Dataset information

Harbor registry scale-ai/swe-atlas-qna
Inspect task scale_ai_swe_atlas_qna
Latest digest sha256:0e26bc0313ae2fc6f912b67b928e648c7f20d17d91f765f702a93042ce5be0e4
Samples 124
Source https://github.com/scaleapi/SWE-Atlas

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