diff --git a/src/transformers/models/qwen3_omni_moe/modeling_qwen3_omni_moe.py b/src/transformers/models/qwen3_omni_moe/modeling_qwen3_omni_moe.py
index 2ddc4d656530..1172ebf90919 100644
--- a/src/transformers/models/qwen3_omni_moe/modeling_qwen3_omni_moe.py
+++ b/src/transformers/models/qwen3_omni_moe/modeling_qwen3_omni_moe.py
@@ -1722,9 +1722,8 @@ def forward(
             past_key_values=past_key_values,
         )
 
-    def _deepstack_process(
-        self, hidden_states: torch.Tensor, visual_pos_masks: torch.Tensor, visual_embeds: torch.Tensor
-    ):
+    def _deepstack_process(self, hidden_states, visual_pos_masks, visual_embeds):
+        visual_pos_masks = visual_pos_masks[..., 0]
         visual_pos_masks = visual_pos_masks.to(hidden_states.device)
         visual_embeds = visual_embeds.to(hidden_states.device, hidden_states.dtype)
         local_this = hidden_states[visual_pos_masks, :].clone() + visual_embeds
@@ -2151,7 +2150,7 @@ def forward(
             use_cache=use_cache,
             output_router_logits=output_router_logits,
             cache_position=cache_position,
-            deepstack_visual_embeds_multiscale=visual_embeds_multiscale,
+            deepstack_visual_embeds=visual_embeds_multiscale,
             visual_pos_masks=visual_pos_masks,
             **kwargs,
         )
diff --git a/src/transformers/models/qwen3_omni_moe/modular_qwen3_omni_moe.py b/src/transformers/models/qwen3_omni_moe/modular_qwen3_omni_moe.py
index 8a7ba792f846..4d1c30f0a4c3 100644
--- a/src/transformers/models/qwen3_omni_moe/modular_qwen3_omni_moe.py
+++ b/src/transformers/models/qwen3_omni_moe/modular_qwen3_omni_moe.py
@@ -1228,6 +1228,10 @@ def __init__(self, config: Qwen3OmniMoeTextConfig):
         )
         self.rotary_emb = Qwen3OmniMoeThinkerTextRotaryEmbedding(config)
 
+    def _deepstack_process(self, hidden_states, visual_pos_masks, visual_embeds):
+        visual_pos_masks = visual_pos_masks[..., 0]
+        return super()._deepstack_process(hidden_states, visual_pos_masks, visual_embeds)
+
 
 @dataclass
 class Qwen3OmniMoeThinkerCausalLMOutputWithPast(MoeCausalLMOutputWithPast):
@@ -1408,7 +1412,7 @@ def forward(
             use_cache=use_cache,
             output_router_logits=output_router_logits,
             cache_position=cache_position,
-            deepstack_visual_embeds_multiscale=visual_embeds_multiscale,
+            deepstack_visual_embeds=visual_embeds_multiscale,
             visual_pos_masks=visual_pos_masks,
             **kwargs,
         )