Piecewise Cuda Graph Support for gpt-oss model (#13045)

Signed-off-by: Oasis-Git <ayw.sirius19@gmail.com>
This commit is contained in:
Yuwei An
2025-11-14 17:28:00 -08:00
committed by GitHub
parent 14dc052382
commit eae59b337e
5 changed files with 28 additions and 24 deletions

View File

@@ -4,6 +4,22 @@ from typing import Any, List, Optional
from sglang.srt.model_executor.forward_batch_info import ForwardBatch
_in_piecewise_cuda_graph = False
def is_in_piecewise_cuda_graph():
return _in_piecewise_cuda_graph
@contextmanager
def enable_piecewise_cuda_graph():
global _in_piecewise_cuda_graph
_in_piecewise_cuda_graph = True
yield
_in_piecewise_cuda_graph = False
@dataclass
class ForwardContext: