fix(quantization): Fix AWQ dequantize on Intel XPU and refactor AutoAWQ config (#42727)
Signed-off-by: Alex <alex.tech.lab@outlook.com> Signed-off-by: AlexHuang <jihuihuang@tencent.com> Co-authored-by: Claude <noreply@anthropic.com> Co-authored-by: Kunshang Ji <kunshang.ji@intel.com> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
This commit is contained in:
co-authored by
Claude
Kunshang Ji
mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
parent
351c72d6e5
commit
8d4f54966c
@@ -48,8 +48,8 @@ from vllm.model_executor.layers.mamba.ops.causal_conv1d import (
|
||||
causal_conv1d_update,
|
||||
)
|
||||
from vllm.model_executor.layers.quantization import QuantizationConfig
|
||||
from vllm.model_executor.layers.quantization.auto_awq import AutoAWQConfig
|
||||
from vllm.model_executor.layers.quantization.auto_gptq import AutoGPTQConfig
|
||||
from vllm.model_executor.layers.quantization.awq_marlin import AWQMarlinConfig
|
||||
from vllm.model_executor.layers.quantization.inc import INCConfig
|
||||
from vllm.model_executor.model_loader.weight_utils import (
|
||||
sharded_weight_loader,
|
||||
@@ -628,7 +628,7 @@ class QwenGatedDeltaNetAttention(GatedDeltaNetAttention):
|
||||
return (
|
||||
current_platform.is_cuda()
|
||||
and not self.gqa_interleaved_layout
|
||||
and isinstance(quant_config, (AWQMarlinConfig, AutoGPTQConfig, INCConfig))
|
||||
and isinstance(quant_config, (AutoAWQConfig, AutoGPTQConfig, INCConfig))
|
||||
)
|
||||
|
||||
def split_ba(self, ba: torch.Tensor) -> tuple[torch.Tensor, torch.Tensor]:
|
||||
|
||||
Reference in New Issue
Block a user