minus-squaretheunknownmuncher@lemmy.worldtoSelfhosted@lemmy.world•Hardware for local inference?linkfedilinkEnglisharrow-up2·2 months agoI use Instinct MI60 GPUs. They are pretty decent performance for local LLM. Connecting multiple computers is going to be impractical because severe bandwidth bottleneck. linkfedilink
minus-squaretheunknownmuncher@lemmy.worldtoSelfhosted@lemmy.world•How many and how much are your subscriptions?linkfedilinkEnglisharrow-up4·2 months ago$0 linkfedilink
minus-squaretheunknownmuncher@lemmy.worldtoSelfhosted@lemmy.world•How would you expose Jellyfin securely without a vpn?linkfedilinkEnglisharrow-up3·edit-22 months agoMeh. If someone hacks my server and watches TV then idc, have at it. They earned it EDIT: the downvotes are warranted but they change nothing linkfedilink
I use Instinct MI60 GPUs. They are pretty decent performance for local LLM. Connecting multiple computers is going to be impractical because severe bandwidth bottleneck.