diff --git a/vllm/config/__init__.py b/vllm/config/__init__.py
index f62e4468ef17..2b6899418376 100644
--- a/vllm/config/__init__.py
+++ b/vllm/config/__init__.py
@@ -3522,6 +3522,10 @@ def __post_init__(self):
                     disable_chunked_prefill_reasons.append(
                         "Only \"last\" pooling supports chunked "
                         "prefill and prefix caching; disabling both.")
+                if not getattr(self.model_config.hf_config, "is_causal", True):
+                    disable_chunked_prefill_reasons.append(
+                        "Only models using causal attention supports chunked "
+                        "prefill and prefix caching; disabling both.")
             elif self.model_config.is_encoder_decoder:
                 self.scheduler_config.max_num_encoder_input_tokens = \
                     MULTIMODAL_REGISTRY.get_encdec_max_encoder_len(self.model_config)