Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Don’t sweat the downvotes.

People have very heated opinions about this technology and any strong statement will have more vote-volatility than usual.

I appreciate you putting a voice to what you said. The historical novelty of what these generative AI’s produce is striking, but the variety of style and form is still narrow enough that you start seeing the signature of CharGPT/SD/etc once that historical novelty wears off. They each have a strong, specific voice and it gets as tiring as any other art style that’s overexposed.



Nobody uses SD base models any more. There are thousands of custom models available now and each of them can be extended even further using LORAs, text inversion, hypernetworks, ControlNet and so on. Your imagination (and your VRAM) is the limit.


You can tell that this is a fast moving space by seeing somewhat outdated commentary even on sites like HN.


It’s ironic that HN commentary is so much like LLM output: usually very good, always authoritative, sometimes wildly wrong.

And when I see the wildly wrong stuff it makes me wonder how many seemingly knowledgeable comments were also wrong, just convincing.

I think a lot of the negativity towards LLMs is because they turn a mirror to our own fallibility in facts and tone.


And here we are debating if the AI is good enough


Lora? That's ancient history. We use LoCons now.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: