hayek@feddit.de to Technology@lemmy.worldEnglish · 1 year agoNvidia released a paper about a 100KB text-to-image model that only trained for 4 minutes but claims to be better than bigger modelsresearch.nvidia.comexternal-linkmessage-square7fedilinkarrow-up198arrow-down16file-textcross-posted to: hackernews@derp.foo
arrow-up192arrow-down1external-linkNvidia released a paper about a 100KB text-to-image model that only trained for 4 minutes but claims to be better than bigger modelsresearch.nvidia.comhayek@feddit.de to Technology@lemmy.worldEnglish · 1 year agomessage-square7fedilinkfile-textcross-posted to: hackernews@derp.foo
minus-squareastrsk@kbin.sociallinkfedilinkarrow-up5·1 year agoThat’s the real meat of this. The future of models will be these smaller, focused “patches” that have some kind of traceable lineage. At least when it comes to marketing and selling these.
That’s the real meat of this. The future of models will be these smaller, focused “patches” that have some kind of traceable lineage. At least when it comes to marketing and selling these.