biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Bugfix] Fix activation quantization for compressed-tensors W4A16 (#31572)

Browse Source 
Signed-off-by: Tmn07 <tmn0796@gmail.com>
This commit is contained in:
Tmn07
2026-01-02 13:13:22 +08:00
committed by GitHub
parent 013b54088c
commit ca179d0f64
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_wNa16.py
View File

@@ -114,7 +114,7 @@ class CompressedTensorsWNA16(CompressedTensorsScheme):
logger.info("Using %s for CompressedTensorsWNA16", kernel_type.__name__)
self._kernel_backends_being_used.add(kernel_type.__name__)
if isinstance(kernel_type, MarlinLinearKernel):
if kernel_type is MarlinLinearKernel:
input_dtype = get_marlin_input_dtype(self.layer_name)
if input_dtype is not None:
mp_linear_kernel_config.act_type = input_dtype