xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-15 01:55:36 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Bug] Fix AttributeError: 'QKVParallelLinear' object has no attribute 'orig_dtype' (#25958)

Browse Source 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
This commit is contained in:
Wentao Ye 2025-09-30 14:13:03 -04:00 committed by GitHub
parent a2e6fa7e03
commit e6a226efba
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a8_fp8.py
View File

@ -66,6 +66,7 @@ class CompressedTensorsW8A8Fp8(CompressedTensorsScheme):
output_size_per_partition = sum(output_partition_sizes) output_size_per_partition = sum(output_partition_sizes)
layer.logical_widths = output_partition_sizes layer.logical_widths = output_partition_sizes
layer.weight_block_size = None layer.weight_block_size = None
layer.orig_dtype = params_dtype
if self.strategy == QuantizationStrategy.BLOCK: if self.strategy == QuantizationStrategy.BLOCK:
assert self.weight_block_size is not None assert self.weight_block_size is not None