Skip to content

[MM][Bugfix] Add MoE verification for multi-modal models #3897

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Nov 4, 2025
+13 −3
Merged

[MM][Bugfix] Add MoE verification for multi-modal models #3897

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
fb67914
Add MoE verification for multi-modal models
shen-shanshan Nov 3, 2025
fe78b82
update var
shen-shanshan Nov 3, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 13 additions & 3 deletions vllm_ascend/utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -672,14 +672,24 @@ def prefill_context_parallel_enable() -> bool:


def is_moe_model(vllm_config: VllmConfig):
"""Checks if the model is a MoE model by config"""
global _IS_MOE_MODEL
if _IS_MOE_MODEL is None:
config = vllm_config.model_config.hf_config
_IS_MOE_MODEL = any('experts' in key.lower()
for key in config.to_dict())
model_configs = vllm_config.model_config.hf_config.to_dict()
_IS_MOE_MODEL = _is_contain_expert(model_configs)
return _IS_MOE_MODEL


def _is_contain_expert(config: Any):
if isinstance(config, dict):
for k, v in config.items():
if "expert" in str(k):
return True
if _is_contain_expert(v):
return True
return False


def weak_ref_tensor(tensor: Any) -> Any:
"""
Create a weak reference to a tensor.
Expand Down
Loading