logoalt Hacker News

0xbadc0de5 • today at 3:02 PM • 1 reply • view on HN

Having used both Sol-6.2 and Astra-6.2 heavily over the last few days (as well as regularly using Claude Opus and Fable) they are marked improvements over the previous releases and comparable to Opus-5.5 and Fable-5.2.

I've stopped trusting any of these influencer-benchmaxxers because they generally don't run any of the workflows of interest to me - it's completely synthetic. The only benchmark that matters is how well any model works for your current use-case.


Replies

adverbly • today at 3:16 PM

No offense, but aren't you wasting your time commenting about that?

It should be obvious that random hacker news comments are worth less than an influencer benchmark, so why waste the time commenting?

What is your goal other than to be an even worse influencer than the very one you're complaining about?

Are you a bot?

➕ show 3 replies