apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-12-24 13:28:13 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

support w4afp8 moe offline permute & load (#5613)

Browse Source
This commit is contained in:
Sunny-bot1
2025-12-22 15:12:57 +08:00
committed by GitHub
parent 81384ef29e
commit 40f3897a4e
3 changed files with 70 additions and 38 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
fastdeploy/model_executor/layers/moe/moe.py
View File

@@ -212,8 +212,10 @@ class FusedMoE(nn.Layer):
self._dtype = self._helper.get_default_dtype()
self.weight_dtype = self._dtype
self.is_quantized = fd_config.model_config.is_quantized and not (
fd_config.quant_config.name() == "mix_quant" and fd_config.quant_config.moe_quant_type is None
self.is_moe_quantized = getattr(self.fd_config.model_config, "is_moe_quantized", False)
self.is_quantized = self.is_moe_quantized or (
fd_config.model_config.is_quantized
and not (fd_config.quant_config.name() == "mix_quant" and fd_config.quant_config.moe_quant_type is None)
)
moe_quant_config = fd_config.quant_config
self.moe_quant_config = moe_quant_config