Will this make the inference more cost effective and reduce the price for r1 model
Thanks for the post, here at SambNova we donβt utilize GPUs for our cloud - We are running on our propriety hardware. With that in mind, we are always striving for new efficiencies and improvements to not only increase metrics like speed and precision but also attempt to lower our costs for all our cloud users. So keep an eye out for upcoming annocuments in the future.
Thanks
Alex
1 Like