Leaklens

SWE-rebench (V1 filtered sample): every task, with its verdict

SWE-rebench, published by Nebius. Paper authors: Ibragim Badertdinov, Alexander Golubev, Maksim Nekrashevich, Anton Shevtsov, Simon Karasik, Andrei Andriushchenko, Maria Trofimova, Daria Litvintseva and Boris Yangel. SWE-rebench (arXiv) · This page covers a random 100-task sample of the V1 filtered split (nebius/SWE-rebench, 6,542 tasks). The sample is the 100 instance IDs with the lowest SHA-256 of "own1:swe-rebench:<instance_id>", so anyone can redraw it. · Report a correction · flags show what the public text and git history contain: public exposure was possible, nothing more.

Download CSV
TaskVerdictIssue title
NCAS-CMS__cf-python-418no overlap found (rule check)Creating time bounds - differing time units causes uncaptured error when collapsing