eicker@lemmy.world to Technology@lemmy.worldEnglish · edit-27 hours agoDeepSeek’s updated V4 Pro AI model struggles on benchmarks, shines in cybersecurity: impresses in niche areas but leaves some developers disappointed in its overall capabilities.www.scmp.comexternal-linkmessage-square1linkfedilinkarrow-up114arrow-down12
arrow-up112arrow-down1external-linkDeepSeek’s updated V4 Pro AI model struggles on benchmarks, shines in cybersecurity: impresses in niche areas but leaves some developers disappointed in its overall capabilities.www.scmp.comeicker@lemmy.world to Technology@lemmy.worldEnglish · edit-27 hours agomessage-square1linkfedilink
minus-squareEager Eagle@lemmy.worldlinkfedilinkEnglisharrow-up4·edit-26 hours agoI’d say that “struggles” is a bit too much. And it’s still 5x to 10x cheaper than the models they compared to in benchmarks, let’s see by how much they’re increasing that cost.
I’d say that “struggles” is a bit too much. And it’s still 5x to 10x cheaper than the models they compared to in benchmarks, let’s see by how much they’re increasing that cost.