EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

fix awq_dequantize (#4333)

Browse Source
This commit is contained in:
Yineng Zhang
2025-03-12 01:04:38 -07:00
committed by GitHub
parent 07f944631e
commit 0a3960f21f
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
sgl-kernel/python/sgl_kernel/gemm.py
View File

@@ -7,7 +7,7 @@ from sgl_kernel.utils import _get_cache_buf, get_cuda_stream
def awq_dequantize( def awq_dequantize(
qweight: torch.Tensor, scales: torch.Tensor, qzeros: torch.Tensor qweight: torch.Tensor, scales: torch.Tensor, qzeros: torch.Tensor
) -> torch.ByteTensor: ) -> torch.ByteTensor:
return torch.ops.sgl_kernels.awq_dequantize(qweight, scales, qzeros) return torch.ops.sgl_kernel.awq_dequantize(qweight, scales, qzeros)
def int8_scaled_mm(mat_a, mat_b, scales_a, scales_b, out_dtype, bias=None): def int8_scaled_mm(mat_a, mat_b, scales_a, scales_b, out_dtype, bias=None):