From 18f0ad76abf44b71070a91caf0ea57458ce05824 Mon Sep 17 00:00:00 2001
From: Xuejun Zhai <Xuejun.Zhai@intel.com>
Date: Sat, 28 Feb 2026 09:59:00 +0800
Subject: [PATCH 1/2] unify api calling

---
 ggml/src/ggml-openvino/utils.cpp | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/ggml/src/ggml-openvino/utils.cpp b/ggml/src/ggml-openvino/utils.cpp
index 05478cbc3e6..cdc1ea04205 100644
--- a/ggml/src/ggml-openvino/utils.cpp
+++ b/ggml/src/ggml-openvino/utils.cpp
@@ -379,9 +379,10 @@ enum ggml_status ov_graph_compute_static(ggml_cgraph * cgraph, std::shared_ptr<o
             }
 
             for (size_t i = 0; i < ov_output_names.size(); i++) {
-                auto * ggml_tensor = ggml_decoder->get_model_outputs().at(ov_output_names[i]);
-                ov::Tensor output_tensor(infer_request->get_output_tensor(i).get_element_type(),
-                                         infer_request->get_output_tensor(i).get_shape(), ggml_tensor->data);
+                auto output_tensor = get_ov_output_tensor(ggml_decoder, ov_output_names[i]);
+                // auto * ggml_tensor = ggml_decoder->get_model_outputs().at(ov_output_names[i]);
+                // ov::Tensor output_tensor(infer_request->get_output_tensor(i).get_element_type(),
+                //                          infer_request->get_output_tensor(i).get_shape(), ggml_tensor->data);
                 infer_request->set_output_tensor(i, output_tensor);
             }
 
@@ -408,9 +409,10 @@ enum ggml_status ov_graph_compute_static(ggml_cgraph * cgraph, std::shared_ptr<o
         }
 
         for (size_t i = 0; i < ov_output_names.size(); i++) {
-            auto * ggml_tensor = ggml_decoder->get_model_outputs().at(ov_output_names[i]);
-            ov::Tensor output_tensor(infer_request->get_output_tensor(i).get_element_type(),
-                                     infer_request->get_output_tensor(i).get_shape(), ggml_tensor->data);
+            auto output_tensor = get_ov_output_tensor(ggml_decoder, ov_output_names[i]);
+            // auto * ggml_tensor = ggml_decoder->get_model_outputs().at(ov_output_names[i]);
+            // ov::Tensor output_tensor(infer_request->get_output_tensor(i).get_element_type(),
+            //                          infer_request->get_output_tensor(i).get_shape(), ggml_tensor->data);
             infer_request->set_output_tensor(i, output_tensor);
         }
 

From 2e025bf073d586559a95c73beb6b7999a936a02a Mon Sep 17 00:00:00 2001
From: Xuejun Zhai <xuejun.zhai@intel.com>
Date: Sat, 28 Feb 2026 09:59:38 +0800
Subject: [PATCH 2/2] Update utils.cpp

---
 ggml/src/ggml-openvino/utils.cpp | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/ggml/src/ggml-openvino/utils.cpp b/ggml/src/ggml-openvino/utils.cpp
index cdc1ea04205..4a735df8a5c 100644
--- a/ggml/src/ggml-openvino/utils.cpp
+++ b/ggml/src/ggml-openvino/utils.cpp
@@ -380,9 +380,6 @@ enum ggml_status ov_graph_compute_static(ggml_cgraph * cgraph, std::shared_ptr<o
 
             for (size_t i = 0; i < ov_output_names.size(); i++) {
                 auto output_tensor = get_ov_output_tensor(ggml_decoder, ov_output_names[i]);
-                // auto * ggml_tensor = ggml_decoder->get_model_outputs().at(ov_output_names[i]);
-                // ov::Tensor output_tensor(infer_request->get_output_tensor(i).get_element_type(),
-                //                          infer_request->get_output_tensor(i).get_shape(), ggml_tensor->data);
                 infer_request->set_output_tensor(i, output_tensor);
             }
 
@@ -410,9 +407,6 @@ enum ggml_status ov_graph_compute_static(ggml_cgraph * cgraph, std::shared_ptr<o
 
         for (size_t i = 0; i < ov_output_names.size(); i++) {
             auto output_tensor = get_ov_output_tensor(ggml_decoder, ov_output_names[i]);
-            // auto * ggml_tensor = ggml_decoder->get_model_outputs().at(ov_output_names[i]);
-            // ov::Tensor output_tensor(infer_request->get_output_tensor(i).get_element_type(),
-            //                          infer_request->get_output_tensor(i).get_shape(), ggml_tensor->data);
             infer_request->set_output_tensor(i, output_tensor);
         }