JOMusic@lemmy.ml to Technology@lemmy.worldEnglish · 2 天前US Bill proposed to jail people who download Deepseekwww.404media.coexternal-linkmessage-square139fedilinkarrow-up1808arrow-down116cross-posted to: technology@beehaw.orgopensource@lemmy.mlnottheonion@lemmy.world
arrow-up1792arrow-down1external-linkUS Bill proposed to jail people who download Deepseekwww.404media.coJOMusic@lemmy.ml to Technology@lemmy.worldEnglish · 2 天前message-square139fedilinkcross-posted to: technology@beehaw.orgopensource@lemmy.mlnottheonion@lemmy.world
minus-squareKnock_Knock_Lemmy_In@lemmy.worldlinkfedilinkEnglisharrow-up2·2 天前It’s easy to run a distilled version of the R1 model locally. It’s very difficult to run the full version. Min $6k to get 7 tokens per second.
minus-squarerumba@lemmy.ziplinkfedilinkEnglisharrow-up2·edit-22 天前Here’s one for 2k if you don’t mine jank (edit: and 3-4 tokens :) ) https://digitalspaceport.com/how-to-run-deepseek-r1-671b-fully-locally-on-2000-epyc-rig/
minus-squareKyuuketsuki@lemmy.mllinkfedilinkEnglisharrow-up1·2 天前I hear its easy, but I’ve had no luck at all on the most distilled models (for prelim testing), and am wondering how things have broken so badly.
It’s easy to run a distilled version of the R1 model locally. It’s very difficult to run the full version. Min $6k to get 7 tokens per second.
Here’s one for 2k if you don’t mine jank (edit: and 3-4 tokens :) )
https://digitalspaceport.com/how-to-run-deepseek-r1-671b-fully-locally-on-2000-epyc-rig/
I hear its easy, but I’ve had no luck at all on the most distilled models (for prelim testing), and am wondering how things have broken so badly.