News for hackers. Stuff that matters.

A comment on Why SWE-bench Verified no longer measures frontier coding capabilities. Parent comment

Re:Why SWE-bench Verified no longer measures frontier coding... by MattRix on Sunday April 26, 2026 @11:38AM (#47911163)
They mention this in the article. This is why private (non public) benchmark tasks that have been made from scratch are necessary.
[ Parent | on HN ]

There are no comments on this comment yet.