AlpinDale fa84f8102e kernels: split marlin kernels for faster compile, fix MoE, temporarily remove HQQ (#1119) 2 settimane fa
..
marlin_kernels fa84f8102e kernels: split marlin kernels for faster compile, fix MoE, temporarily remove HQQ (#1119) 2 settimane fa
align_block_size_kernel.cu f1d0b77c92 [0.6.0] Release Candidate (#481) 5 mesi fa
marlin_moe_ops.cu fa84f8102e kernels: split marlin kernels for faster compile, fix MoE, temporarily remove HQQ (#1119) 2 settimane fa
marlin_moe_ops.h fa84f8102e kernels: split marlin kernels for faster compile, fix MoE, temporarily remove HQQ (#1119) 2 settimane fa
moe_ops.h f1d0b77c92 [0.6.0] Release Candidate (#481) 5 mesi fa
topk_softmax_kernels.cu fa84f8102e kernels: split marlin kernels for faster compile, fix MoE, temporarily remove HQQ (#1119) 2 settimane fa
torch_bindings.cpp fa84f8102e kernels: split marlin kernels for faster compile, fix MoE, temporarily remove HQQ (#1119) 2 settimane fa