EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

[NVIDIA] Fix num_experts in modelopt_quant (#8811)

Browse Source
This commit is contained in:
Shu Wang
2025-08-06 16:35:07 -05:00
committed by GitHub
parent 01c99a9959
commit 288ae41f7a
2 changed files with 7 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
python/sglang/srt/layers/quantization/modelopt_quant.py
View File

@@ -764,8 +764,6 @@ class ModelOptNvFp4FusedMoEMethod(FusedMoEMethodBase):
)
# TODO(ch-wan): check if this is needed
layer.num_experts = num_experts
layer.num_local_experts = num_experts
layer.intermediate_size_per_partition = intermediate_size_per_partition
layer.params_dtype = params_dtype
layer.quant_config = self.quant_config
@@ -1106,7 +1104,7 @@ class ModelOptNvFp4FusedMoEMethod(FusedMoEMethodBase):
layer.w13_weight_scale,
)
print("Applied flashinfer weight processing for both w13 and w2")
logger.info_once("Applied flashinfer weight processing for both w13 and w2")
else:
# CUTLASS processing - handle w13 and w2 separately
@@ -1126,7 +1124,7 @@ class ModelOptNvFp4FusedMoEMethod(FusedMoEMethodBase):
layer.w2_weight = Parameter(layer.w2_weight.data, requires_grad=False)
# Both flashinfer cutlass and regular cutlass use same processing for w2
print("Applied weight processing for both w13 and w2")
logger.info_once("Applied weight processing for both w13 and w2")
# Set up CUTLASS MoE parameters
device = layer.w13_weight.device