Remix.run Logo
throw1234567891 2 days ago

You don’t need a datacenter. 1T model requires 1T of RAM. Not easy to come by today in a compact factor but nowhere near the datacenter.

nullc 21 hours ago | parent [-]

> 1T model requires 1T of RAM

assuming 8 bits per weight, but 4bits is commonly used once you get to that scale.