[BugFix] Disable enable_shared_expert_dp by default if tensor_parallel_size=1 (#6361)
### What this PR does / why we need it? Disable enable_shared_expert_dp by default if tensor_parallel_size=1 - vLLM version: v0.14.1 - vLLM main: https://github.com/vllm-project/vllm/commit/dc917cceb877dfd13f98c538c4c96158047d98bd Signed-off-by: underfituu <hzhucong@163.com>
This commit is contained in:
1 parent
8b0a7b6d80
commit
df588ed488
1 file changed
+1
@@ -61,6 +61,7 @@ class AscendConfig:
|
||||
self.enable_shared_expert_dp = (
|
||||
additional_config.get("enable_shared_expert_dp", False)
|
||||
and vllm_config.parallel_config.enable_expert_parallel
|
||||
and vllm_config.parallel_config.tensor_parallel_size > 1
|
||||
)
|
||||
from vllm_ascend.utils import enable_sp
|
||||
|
||||
|
||||
Reference in new issue
Block a user