Grok is my go-to LLM for research not because I care for Elon Musk but because it's really good. I'm not even sure if I'm paying for it? I have an X Premium account but I don't know if that's giving me more Grok usage or what. Either way, its research mode is very good and frequently pulls off tasks the competitors struggle with in my experience.
This is much easier fixed by judges actually locking up career criminals but they're so "empathetic" they let lunatics run freely and terrorize communities.
HTML and CSS are not a separation of concerns in anything I've ever built. They're both representing the layout/design. I don't want them separate.
You end up having to keep them synchronized anyway, messing with your HTML is frequently gonna break your CSS. Tailwind is infinitely more pleasant to write than a separate CSS file that has to be kept in sync with the HTML.
Google Search's AI has gotten... better. But it's still frequently spouting as much made up shit as I expected from LLMs not doing their research a couple years ago.
Just this morning I looked up what time today's Nintendo Direct livestream is. It correctly told me it's today at 7am PT (confirmed from actual link sources). It then tried to get smart and tell me what that would be in my city's timezone and it was an hour off (possibly because I'm in a timezone without daylight savings)
Yes it's definitely not SOTA, and is worse than all the chatbots I use. But it's correct frequently enough that the convenience + speed does it for me.
> Best would be, if there was a model in the first place that would just choose the most optimal model to execute your request, otherwise why bother automating everything in the first place?
Practically every chat UI these days defaults to "auto" mode which does that
reply