xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-10 03:05:02 +08:00
Code Issues Packages Projects Releases Wiki Activity

[config] Expose get_total_num_hidden_layers() in ModelConfig (#28961)

Browse Source 
Signed-off-by: tovam <tovam@pliops.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
This commit is contained in:
Tova Movshovitz 2025-11-19 13:37:11 +02:00 committed by GitHub
parent 97cfa99d59
commit ba558c029a
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 10 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
vllm/config/model.py
View File

@ -1369,11 +1369,7 @@ class ModelConfig:
# Coerce to 0 if explicitly set to None
return num_experts or 0
def get_layers_start_end_indices(
self, parallel_config: ParallelConfig
) -> tuple[int, int]:
from vllm.distributed.utils import get_pp_indices
def get_total_num_hidden_layers(self) -> int:
if (
self.hf_text_config.model_type == "deepseek_mtp"
or self.hf_config.model_type == "mimo_mtp"
@ -1393,6 +1389,15 @@ class ModelConfig:
total_num_hidden_layers = getattr(
self.hf_text_config, "num_hidden_layers", 0
)
return total_num_hidden_layers
def get_layers_start_end_indices(
self, parallel_config: ParallelConfig
) -> tuple[int, int]:
from vllm.distributed.utils import get_pp_indices
total_num_hidden_layers = self.get_total_num_hidden_layers()
# the layout order is: DP x PP x TP
pp_rank = (
parallel_config.rank // parallel_config.tensor_parallel_size