diff --git a/ggml/src/ggml-rknpu2/ggml-rknpu2.cpp b/ggml/src/ggml-rknpu2/ggml-rknpu2.cpp
index 1f7ac3d5f..108446104 100644
--- a/ggml/src/ggml-rknpu2/ggml-rknpu2.cpp
+++ b/ggml/src/ggml-rknpu2/ggml-rknpu2.cpp
@@ -958,6 +958,9 @@ static void dequantize_row(
     } else if (tensor->type == GGML_TYPE_Q4_0) {
         const block_q4_0* src = (const block_q4_0*)raw_data;
         dequantize_row_q4_0(src + (size_t)n * (K / QK4_0), row_out, K);
+    } else if (tensor->type == GGML_TYPE_Q4_K) {
+        const block_q4_K* src = (const block_q4_K*)raw_data;
+        dequantize_row_q4_K(src + (size_t)n * (K / QK_K), row_out, K);
     } else {
         GGML_ASSERT(false && "Unsupported weight type for NPU pipeline");
     }
diff --git a/ggml/src/ggml-rknpu2/rknpu2-configuration.cpp b/ggml/src/ggml-rknpu2/rknpu2-configuration.cpp
index 1b659081f..70cedfe16 100644
--- a/ggml/src/ggml-rknpu2/rknpu2-configuration.cpp
+++ b/ggml/src/ggml-rknpu2/rknpu2-configuration.cpp
@@ -212,6 +212,7 @@ Rknpu2ConfigManager::Rknpu2ConfigManager() {
     rk3588_config.default_patterns[(int)GGML_TYPE_Q8_0] = {"W8A8_STANDARD"};
     rk3588_config.default_patterns[(int)GGML_TYPE_Q6_K] = {"W8A8_STANDARD", "W4A4_HADAMARD"};
     rk3588_config.default_patterns[(int)GGML_TYPE_Q4_0] = {"W4A4_HADAMARD"};
+    rk3588_config.default_patterns[(int)GGML_TYPE_Q4_K] = {"W4A4_HADAMARD"};
 
     device_configs["RK3588"] = rk3588_config;