diff --git a/vllm/models/qwen4_exp/amd/mtp.py b/vllm/models/qwen4_exp/amd/mtp.py
index a2b420603f..d103f17d0a 100644
--- a/vllm/models/qwen4_exp/amd/mtp.py
+++ b/vllm/models/qwen4_exp/amd/mtp.py
@@ -33,7 +33,7 @@ from vllm.model_executor.layers.vocab_parallel_embedding import (
     VocabParallelEmbedding,
 )
 from vllm.model_executor.model_loader.utils import configure_quant_config
-from vllm.model_executor.models.interfaces import SupportsPP
+from vllm.model_executor.models.interfaces import LocalArgmaxMixin, SupportsPP
 from vllm.model_executor.models.qwen3_5 import Qwen3_5Model
 from vllm.model_executor.models.utils import (
     AutoWeightsLoader,
@@ -379,7 +379,7 @@ class Qwen4ExpMultiTokenPredictor(nn.Module):
         "hidden_states": 0,
     }
 )
-class Qwen4ExpMTP(nn.Module, SupportsPP, Qwen4ExpMixtureOfExperts):
+class Qwen4ExpMTP(LocalArgmaxMixin, nn.Module, SupportsPP, Qwen4ExpMixtureOfExperts):
     packed_modules_mapping = {
         "qkv_proj": ["q_proj", "k_proj", "v_proj"],
         "gate_up_proj": ["gate_proj", "up_proj"],
diff --git a/vllm/models/qwen4_exp/nvidia/mtp.py b/vllm/models/qwen4_exp/nvidia/mtp.py
index af3bcd4659..1fbfb0b25c 100644
--- a/vllm/models/qwen4_exp/nvidia/mtp.py
+++ b/vllm/models/qwen4_exp/nvidia/mtp.py
@@ -32,7 +32,7 @@ from vllm.model_executor.layers.vocab_parallel_embedding import (
     VocabParallelEmbedding,
 )
 from vllm.model_executor.model_loader.utils import configure_quant_config
-from vllm.model_executor.models.interfaces import SupportsPP
+from vllm.model_executor.models.interfaces import LocalArgmaxMixin, SupportsPP
 from vllm.model_executor.models.qwen3_5 import Qwen3_5Model
 from vllm.model_executor.models.utils import (
     AutoWeightsLoader,
@@ -359,7 +359,7 @@ class Qwen4ExpMultiTokenPredictor(nn.Module):
         return loader.load_weights(weights, mapper=mapper)
 
 
-class Qwen4ExpMTP(nn.Module, SupportsPP, Qwen4ExpMixtureOfExperts):
+class Qwen4ExpMTP(LocalArgmaxMixin, nn.Module, SupportsPP, Qwen4ExpMixtureOfExperts):
     packed_modules_mapping = {
         "qkv_proj": ["q_proj", "k_proj", "v_proj"],
         "gate_up_proj": ["gate_proj", "up_proj"],