apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-05 08:37:06 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[Feature] support custom all-reduce (#2758)

Browse Source 
* [Feature] support custom all-reduce

* add vllm adapted
This commit is contained in:
zhink
2025-07-09 16:00:27 +08:00
committed by GitHub
parent be21ef5047
commit b89180f1cd
16 changed files with 1194 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
fastdeploy/engine/config.py
View File

@@ -447,6 +447,7 @@ class ParallelConfig:
tensor_parallel_size: int = 1,
data_parallel_size: int = 1,
enable_expert_parallel: bool = False,
enable_custom_all_reduce: bool = False,
):
"""
Initialize the ParallelConfig class.
@@ -462,6 +463,7 @@ class ParallelConfig:
self.enable_expert_parallel = enable_expert_parallel
self.expert_parallel_size = data_parallel_size
self.local_data_parallel_id = 0
self.enable_custom_all_reduce = enable_custom_all_reduce
def print(self):
"""
@@ -587,6 +589,7 @@ class Config:
max_capture_batch_size: int = 64,
guided_decoding_backend: Optional[str] = None,
disable_any_whitespace: bool = False,
enable_custom_all_reduce: bool = False,
):
"""
Initialize the Config class.