Autor | SHA1 Mensaxe | Data |
---|---|---|
|
ed63c079f7 Triton: remove atomic add op from awq triton (#1094) | hai 3 semanas |
|
cbde3c66a5 quants: improve awq_triton throughput (#998) | hai 1 mes |
|
fcfcfc65e1 quants: add triton kernels for AWQ (#946) | hai 1 mes |