logoalt Hacker News

aesthesia • yesterday at 4:57 PM • 2 replies • view on HN

Approximately every benchmark has significant issues.


Replies

mathis-l • yesterday at 5:02 PM

Of course. That’s why I opened the issue so that the team can improve it.

agentdev001 • yesterday at 5:26 PM

Every existing, and not yet existing benchmark.