Using CUDA to accelerate inference results on LM STUDIO will display garbled characters, while pure CPU inference is normal

#1
by zhangcaohai - opened

Using CUDA to accelerate inference results on LM STUDIO will display garbled characters, while pure CPU inference is normal

UD-Q4_K_XL

Display mai-ds-r1-ud
8A69!AA7.;E8:"-"E;;6%49327

Sign up or log in to comment