Comment on DeepSeek Coder: Let the Code Write ItselfparentComments−findjashua2yLM Studio is the easiest way to do it−Sabinus2yThat's what I've been playing with. I can load 9 layers of a mixtral descendant into the 12gb vram for GPU and the rest into ~28gb ram for the CPU to work on. It chugs the system sometimes but the models are interestingly capable.
Comments
LM Studio is the easiest way to do it
That's what I've been playing with. I can load 9 layers of a mixtral descendant into the 12gb vram for GPU and the rest into ~28gb ram for the CPU to work on. It chugs the system sometimes but the models are interestingly capable.