diff --git a/vllm/attention/ops/prefix_prefill.py b/vllm/attention/ops/prefix_prefill.py
index fbb6757ee3043..5fca1639363e0 100644
--- a/vllm/attention/ops/prefix_prefill.py
+++ b/vllm/attention/ops/prefix_prefill.py
@@ -11,7 +11,7 @@ from vllm.platforms import current_platform
 
 # Static kernels parameters
 BASE_BLOCK = 128 if current_platform.has_device_capability(80) else 64
-NUM_WARPS = 8
+NUM_WARPS = 4 if current_platform.is_rocm() else 8
 
 # To check compatibility
 IS_TURING = current_platform.get_device_capability() == (7, 5)