obscura/vllm
2
0
Fork 0
mirror of https://github.com/vllm-project/vllm.git synced 2026-06-06 00:16:14 +00:00
Code Issues Packages Projects Releases Wiki Activity

[XPU] [Bug] remove xpuw4a16 output size check (#44168)

Browse Source 
Signed-off-by: Zhu, Zufang <zufang.zhu@intel.com>
Co-authored-by: Kunshang Ji <kunshang.ji@intel.com>
This commit is contained in:
zofia
2026-06-02 22:26:20 +08:00
committed by GitHub
parent c91a87f01a
commit 6314de8bad
1 changed files with 0 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files
vllm/model_executor/kernels/linear/mixed_precision/xpu.py
-7
View File
@@ -51,13 +51,6 @@ class XPUwNa16LinearKernel(MPLinearKernel):
"XPUwNa16, supported sizes are multiples of 32",
)
if c.partition_weight_shape[1] % 32 != 0:
return (
False,
f"Output size ({c.partition_weight_shape[1]}) not supported by "
"XPUWNA16, supported sizes are multiples of 32",
)
return True, None
def process_weights_after_loading(self, layer: torch.nn.Module):