vrr/kvcache-ai-ktransformers
2
0
Fork 0
mirror of https://github.com/kvcache-ai/ktransformers.git synced 2025-09-12 08:09:42 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 2

GLM4 and SmallThinker

Browse source
This commit is contained in:
qiyuxinlin 2025-07-25 16:56:36 +00:00
parent c7307aa0ae
commit 9e1560bb82
7 changed files with 58 additions and 37 deletions
Download patch file Download diff file Expand all files Collapse all files

6
ktransformers/operators/experts.py
View file

@ -194,7 +194,7 @@ class KExpertsCPU(KExpertsBase):
64,
10,
1024,
self.config.model_type != "smallthinker",
self.config.hidden_act == 'silu',
gate_ptr,
up_ptr,
down_ptr,
@ -215,7 +215,7 @@ class KExpertsCPU(KExpertsBase):
self.config.hidden_size,
self.config.moe_intermediate_size,
max(cuda_graphs) if isinstance(cuda_graphs, list) else Config().chunk_size,
self.config.model_type != "smallthinker",
self.config.hidden_act == 'silu',
gate_ptr,
up_ptr,
down_ptr,
@ -234,7 +234,7 @@ class KExpertsCPU(KExpertsBase):
self.config.hidden_size,
self.config.moe_intermediate_size,
max(cuda_graphs) if isinstance(cuda_graphs, list) else Config().chunk_size,
self.config.model_type != "smallthinker",
self.config.hidden_act == 'silu',
gate_ptr,
up_ptr,
down_ptr,