diff --git a/vllm/v1/attention/backends/gdn_attn.py b/vllm/v1/attention/backends/gdn_attn.py
index 16a2c39176..6b99ec5452 100644
--- a/vllm/v1/attention/backends/gdn_attn.py
+++ b/vllm/v1/attention/backends/gdn_attn.py
@@ -546,7 +546,10 @@ class GDNAttentionMetadataBuilder(AttentionMetadataBuilder[GDNAttentionMetadata]
                         cu_seqlens=prefill_query_start_loc,
                     )
                 )
-            else:
+            elif self.gdn_prefill_backend != "flashinfer":
+                # FlashInfer's chunk kernel consumes cu_seqlens directly and
+                # never reads chunk_indices/chunk_offsets, so skip the FLA
+                # metadata (CPU compute + H2D copies) entirely.
                 chunk_indices, chunk_offsets = self._build_chunk_metadata(
                     prefill_query_start_loc,
                     prefill_query_start_loc_cpu,