inari@piefed.zip to Technology@lemmy.worldEnglish · edit-24 months agoDeepSeek ditches Nvidia for Huawei chips in V4 launchcybernews.comexternal-linkmessage-square84linkfedilinkarrow-up1292arrow-down15
arrow-up1287arrow-down1external-linkDeepSeek ditches Nvidia for Huawei chips in V4 launchcybernews.cominari@piefed.zip to Technology@lemmy.worldEnglish · edit-24 months agomessage-square84linkfedilink
minus-squarebrucethemoose@lemmy.worldlinkfedilinkEnglisharrow-up5arrow-down1·4 months agoI just meant for mass inference serving. Yeah, I haven’t seen much in the way of bitnet training savings yet, like regular old QAT. It does appear that Deepseek is finetuning their MoEs in a 4-bit format now, though.
I just meant for mass inference serving.
Yeah, I haven’t seen much in the way of bitnet training savings yet, like regular old QAT. It does appear that Deepseek is finetuning their MoEs in a 4-bit format now, though.