diff --git a/vllm/model_executor/layers/attention/mla_attention.py b/vllm/model_executor/layers/attention/mla_attention.py
index 862f84939853..4859af43ae41 100644
--- a/vllm/model_executor/layers/attention/mla_attention.py
+++ b/vllm/model_executor/layers/attention/mla_attention.py
@@ -530,7 +530,7 @@ def forward_impl(
                 scale=self._k_scale,
             )
 
-        if fp8_attention:
+        if fp8_attention and self.kv_cache_dtype != "fp8_ds_mla":
             kv_cache = kv_cache.view(current_platform.fp8_dtype())
 
         # Sparse MLA impls only support forward_mqa (decode-style attention)
@@ -614,7 +614,7 @@ def forward_impl(
                 # Convert from (N, B, L) to (B, N, L)
                 mqa_ql_nope = mqa_ql_nope.transpose(0, 1)
 
-            if fp8_attention:
+            if fp8_attention and self.impl.supports_quant_query_input:
                 assert mqa_ql_nope.shape[0] == mqa_q_pe.shape[0]
                 assert mqa_ql_nope.shape[1] == mqa_q_pe.shape[1]
                 mqa_q = self._decode_concat_quant_fp8_op(
@@ -1885,6 +1885,8 @@ def __init__(
         self.indexer = indexer
         self.q_pad_num_heads = q_pad_num_heads
 
+        self.supports_quant_query_input = True
+
         # Use flashinfer's optimized concat_mla_k kernel when available.
         # The kernel is optimized for DeepSeek V3 dimensions:
         # num_heads=128, nope_dim=128, rope_dim=64