But they conveniently leave out that it costs money to do anything with AI. It’s more like “open to anyone with a credit card.” The vast majority of people don’t have computers powerful enough to run generative AI models locally, and even then, server farms with a billion GPUs will always produce better results
This means that people have to rely on corporate platforms where you buy tokens that you use to get pulls at the various AI slop slot machines, hoping you get something decent. The mechanics more closely resemble a gacha game than any kind of artistic process
By contrast, learning how to draw, animate or make 3D models costs nothing. There’s free tutorials and tools everywhere, and you can also just pirate commercial ones if you want
i mean, it runs on laptops, its interesting to wrangle it with control nets to try to approach what you want, not art tho, more like visual shorthand builder
If it’s that easy to run I might mess around with it myself 🤔
if you have nvidia, invoke works fineish for more involved shenanigans (open models are definitely shit for real people and stuff people do with like text controlled cloth change with big models), but they can do fine with taking a sketch, filling it out 20 different ways and then a little bit iterating (although i would say, for background foreground photorealistic type picture, ai look is kinda unavoidable, easier to fuck around in photoshop to fix it) or giving ideas.
although it’s not satisfying as sketching with pen on paper, imma be real, just entertaining as riding a bike or something, after 2 months of doing shenanigans myself when bored, i grow kinda tired of messing with it
rip. amd on linux has some support, but on windows it’s fucking shit. or if you mean only cpu - it’s so slow and energy-intensive, it’s not worth it i think? there were some moves to try to speed it up on cpus, but i haven’t looked into the results
Is an AMD GPU + Windows a complete non-starter then? That’s a shame
i think they have something called amuse, but it’s very bare program to run prompts, without interesting stuff like control nets or at least was.
and there is some shit to swap cuda for amd supported thingy, but it’s so much headache, it’s not worth it, at least i think so
No it’s not. You use direct ml, it’s slower but does work. Last I heard “shark” works? But I haven’t looked at it in ages. There are options though.