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 MagicMoonlight on Sunday April 26, 2026 @09:47PM (#47916844)
This article says anthropic models can write out the entire benchmark solution set word for word from memory
[ Parent | on HN ]

There are no comments on this comment yet.