Commit 12cdfda7 authored by zhuwenwen's avatar zhuwenwen
Browse files

pd separation uses default scheduling

parent c5980399
......@@ -1089,6 +1089,8 @@ class Scheduler(SchedulerInterface):
def schedule(self) -> SchedulerOutput:
if envs.VLLM_USE_PD_SPLIT:
if self.connector is not None:
return self.schedule_default()
if self.use_mla:
if self.full_cuda_graph and self.num_spec_tokens > 0:
return self.schedule_split_pd()
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment