AlpinDale fa84f8102e kernels: split marlin kernels for faster compile, fix MoE, temporarily remove HQQ (#1119) 5 days ago
..
awq_marlin_repack.cu a113309876 kernel: add meta functions for ops to prevent graph breaks (#1019) 1 month ago
gptq_marlin.cu fa84f8102e kernels: split marlin kernels for faster compile, fix MoE, temporarily remove HQQ (#1119) 5 days ago
gptq_marlin_repack.cu fa84f8102e kernels: split marlin kernels for faster compile, fix MoE, temporarily remove HQQ (#1119) 5 days ago
marlin.cuh fa84f8102e kernels: split marlin kernels for faster compile, fix MoE, temporarily remove HQQ (#1119) 5 days ago
marlin_dtypes.cuh fa84f8102e kernels: split marlin kernels for faster compile, fix MoE, temporarily remove HQQ (#1119) 5 days ago