Blog post · 2026

SWE-Bench Pro V2

Scale AI

URLhttps://labs.scale.com/blog/swe-bench-pro-v2
TopicsReplicHAI / Agent Evals / 02 Software engineering & coding
Tagsagent-eval

Abstract

Measures: Repaired SWE-bench Pro public set (642 tasks). Metric: % resolved.

BibTeX

@misc{scale_ai_swe-bench_2026,
	title = {{SWE}-{Bench} {Pro} {V2}},
	url = {https://labs.scale.com/blog/swe-bench-pro-v2},
	abstract = {Measures: Repaired SWE-bench Pro public set (642 tasks). Metric: \% resolved.},
	author = {{Scale AI}},
	month = sep,
	year = {2026},
	keywords = {agent-eval},
}