Trilogy3452@lemmy.world to Technology@lemmy.worldEnglish · 16 days agoAnthropic/OpenAI may be spending more than $1000 for every $100 you pay themea.rna.nlexternal-linkmessage-square178linkfedilinkarrow-up1687arrow-down111cross-posted to: hackernews@lemmy.bestiver.se
arrow-up1676arrow-down1external-linkAnthropic/OpenAI may be spending more than $1000 for every $100 you pay themea.rna.nlTrilogy3452@lemmy.world to Technology@lemmy.worldEnglish · 16 days agomessage-square178linkfedilinkcross-posted to: hackernews@lemmy.bestiver.se
minus-squareShteou@lemmy.worldlinkfedilinkEnglisharrow-up3·16 days agoIt is sorta. Training is orders of magnitudes more intensive than inference, but we infer billions of times within a model generation.
It is sorta. Training is orders of magnitudes more intensive than inference, but we infer billions of times within a model generation.