Best way to offload layers on cPU/RAM for 6x24gb GPU
#4
by
djdeniro
- opened
Hello! For 256 RAM + 144 VRAM what is best way to offload layers to CPU?
For Q2_K_XL and Q4_K_XL.
Thanks!
I wrote about on how to offload more layers for more performance here: https://docs.unsloth.ai/basics/qwen3-coder#improving-generation-speed :)