Skip to content

Commit 82db2f9

Browse files
jiqing-fengSunMarc
authored andcommitted
check shift
Signed-off-by: jiqing-feng <jiqing.feng@intel.com>
1 parent b621b41 commit 82db2f9

File tree

1 file changed

+1
-0
lines changed

1 file changed

+1
-0
lines changed

optimum/quanto/tensor/weights/qbits.py

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -110,6 +110,7 @@ def create(qtype, axis, group_size, size, stride, data, scale, shift, requires_g
110110
and group_size == 128
111111
and len(size) == 2
112112
and data.device.type == "xpu"
113+
and shift.dtype == torch.int8
113114
and version.parse(torch.__version__).release >= version.parse("2.8.0").release
114115
):
115116
if type(data) is PackedTensor:

0 commit comments

Comments
 (0)