@Liathetrader It’s an LLM, could be hallucinating so take it with a grain of salt, that being said they almost certainly trained it at bare minimum on H100s
@RedPandaMining@BitsBeTrippin@pmarca@lexfridman R1 isn’t the model you want for coding v3 is and it’s MUCH MUCH larger, think hundreds of gigs of vram if u want it to be on gpus.
R1 can run on a mid level gpu pretty easily though.
And besides all that deepseek pricing is cheaper than vast :/