diff --git a/vllm/model_executor/models/bailing_moe_linear.py b/vllm/model_executor/models/bailing_moe_linear.py
index 55ea1bad44db..bcf1200bd1b4 100644
--- a/vllm/model_executor/models/bailing_moe_linear.py
+++ b/vllm/model_executor/models/bailing_moe_linear.py
@@ -205,13 +205,19 @@ def __init__(
             self.q_a_layernorm = None
             self.q_b_proj = None
 
-        rope_parameters = _build_rope_parameters(config)
+        rope_parameters = _build_rope_parameters(config) or {}
+        # MLA rotates the full qk_rope_head_dim,
+        # partial_rotary_factor is for the linear-attn head only.
+        rope_parameters = {
+            k: v for k, v in rope_parameters.items() if k != "partial_rotary_factor"
+        }
+        rope_parameters["rope_dim"] = self.qk_rope_head_dim
         max_position = getattr(config, "max_position_embeddings", 8192)
         self.rotary_emb = get_rope(
             head_size=self.qk_rope_head_dim,
             max_position=max_position,
             is_neox_style=False,
-            rope_parameters=rope_parameters or None,
+            rope_parameters=rope_parameters,
         )
 
         # Build MLAModules for MultiHeadLatentAttentionWrapper