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

GLM-4.5 and GLM-4.5-Air both support (#8804)

Browse Source
This commit is contained in:
Yuxuan Zhang
2025-08-05 18:29:20 +08:00
committed by GitHub
parent 40e3b2beeb
commit a4b0d5c9e5
1 changed files with 1 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
python/sglang/srt/models/glm4_moe.py
View File

@@ -785,7 +785,7 @@ class Glm4MoeForCausalLM(DeepseekV2ForCausalLM):
) )
def determine_num_fused_shared_experts( def determine_num_fused_shared_experts(
self, architecture: str = "DeepseekV3ForCausalLM" self, architecture: str = "Glm4MoeForCausalLM"
): ):
self.num_fused_shared_experts = 0 self.num_fused_shared_experts = 0
if global_server_args_dict["disable_shared_experts_fusion"]: if global_server_args_dict["disable_shared_experts_fusion"]:
@@ -797,7 +797,6 @@ class Glm4MoeForCausalLM(DeepseekV2ForCausalLM):
not _is_cuda not _is_cuda
or torch.cuda.get_device_capability("cuda") < (8, 0) or torch.cuda.get_device_capability("cuda") < (8, 0)
or self.config.architectures[0] != architecture or self.config.architectures[0] != architecture
or self.config.n_routed_experts != 128
or self.config.n_shared_experts != 1 or self.config.n_shared_experts != 1
): ):
disable_reason = "Only GLM-4.5 on NV-platform with capability >= 80 can use shared experts fusion optimization." disable_reason = "Only GLM-4.5 on NV-platform with capability >= 80 can use shared experts fusion optimization."