Hacker News new | ask | show | jobs
by liuliu 81 days ago
It needs a mlx fork because the lowest bit in mlx is 2 currently (for affine quantization).
1 comments
riidom 80 days ago
That mlx is for apple hardware only, though? Or did I misunderstand something.
link
dragonwriter 80 days ago
It needs a llama.cpp fork, too; so the stock runtime (based on stock llama.cpp) used by LM Studio presumably won't work for it.
link