News for hackers. Stuff that matters.

A comment on Claude Sonnet 5. Parent comment

Re:Claude Sonnet 5 by energy123 on Tuesday June 30, 2026 @02:40PM (#48737340)
The arguable caveat is Sonnet may run faster (although this isn't known for sure, due to more tokens being used for the same task), so you can potentially get more done in a synchronous iterative workflow

I don't really believe this however, because so much time is spent fixing up after models, that a slower but more intelligent model is a net time saver in my experience.

[ Parent | on HN ]
Re:Claude Sonnet 5 by kolinko on Tuesday June 30, 2026 @05:39PM (#48739570)
From my benchmarks, sadly, it doesn't seem to be the case much. Surprisingly. I found Sonnet comparable in speed to Opus (sic), but perhaps I was testing it wrong?
[ Parent | on HN ]
Re:Claude Sonnet 5 by riverbirch on Tuesday June 30, 2026 @06:38PM (#48740180)
I can confirm this, I too I'm not seeing much of a difference in practice
[ Parent | on HN ]
Re:Claude Sonnet 5 by usef- on Tuesday June 30, 2026 @11:42PM (#48741993)
Given the chip shortage, I wonder if they push Sonnet to the older chips. It presumably requires less memory.
[ Parent | on HN ]