[DeepSeek] Use torch.mm for bf16xbf16->fp32 gemm (#41300)
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
This commit is contained in:
@@ -304,13 +304,6 @@ def cpu_unquantized_gemm(
|
||||
return layer.cpu_linear(x, weight, bias)
|
||||
|
||||
|
||||
def cublas_gemm_bf16_bf16_fp32(
|
||||
x: torch.Tensor,
|
||||
weight: torch.Tensor,
|
||||
):
|
||||
return ops.router_gemm_bf16_fp32(x, weight)
|
||||
|
||||
|
||||
def dispatch_unquantized_gemm() -> Callable[..., torch.Tensor]:
|
||||
if current_platform.is_rocm():
|
||||
return rocm_unquantized_gemm
|
||||
|
||||
Reference in New Issue
Block a user