DeepSWE v1

A fully matched harness comparison: the same 113 DeepSWE v1 tasks, GPT-5.6 Sol, high reasoning effort, k=1. Only the harness changes.

Leaderboard

Official post-hoc scoring at the 90 minute cutoff. Our bar is green.

Per-task outcomes

jcode win codex win both pass both fail

Agent runtime

jcode passed jcode failed

Per-task breakdown

Click a column to sort. Filters narrow to the tasks where the harnesses disagreed.

Task Outcome jcode Codex CLI jcode time Transcripts

Methodology and provenance

Everything needed to reproduce or audit the comparison. Hashes are SHA-256.

Version boundary. This is DeepSWE v1. The current DataCurve leaderboard uses DeepSWE v1.1, so those scores are not directly comparable.

Download the audited leaderboard data