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

fix GLM4_MOE launch with compressed_tensor quant model (#8456)

Browse Source
This commit is contained in:
Minglei Zhu
2025-07-28 01:31:20 -07:00
committed by GitHub
parent 581e7dcb92
commit 25f73c6cf3
1 changed files with 1 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
python/sglang/srt/models/glm4_moe.py
View File

@@ -795,6 +795,7 @@ class Glm4MoeForCausalLM(DeepseekV2ForCausalLM):
elif ( elif (
self.quant_config.get_name() == "fp8" self.quant_config.get_name() == "fp8"
or self.quant_config.get_name() == "blockwise_int8" or self.quant_config.get_name() == "blockwise_int8"
or self.quant_config.get_name() == "compressed_tensors"
): ):
suffix_list = [ suffix_list = [
"down_proj.weight", "down_proj.weight",