[OpenVINO][Draft]add support for dinov3

optimum/exporters/openvino/model_configs.py

@@ -4665,6 +4665,25 @@ def patch_model_for_export(
         return OVDecoderModelPatcher(self, model, model_kwargs=model_kwargs)
 
 
+@register_in_tasks_manager("dinov3_vit", *["feature-extraction"], library_name="transformers")
+@register_in_tasks_manager("dinov3_convnext", *["feature-extraction"], library_name="transformers")
+class DinoV3OpenVINOConfig(VisionOnnxConfig):
+    MIN_TRANSFORMERS_VERSION = "4.55.0"
+
+
+    NORMALIZED_CONFIG_CLASS = NormalizedVisionConfig
+    @property
+    def inputs(self) -> Dict[str, Dict[int, str]]:
+        return {
+            "pixel_values": {0: "image_batch_size", 1: "num_channels", 2: "height", 3: "width"},
+        }
+
+    @property
+    def outputs(self) -> Dict[str, Dict[int, str]]:
+        return {
+            "last_hidden_state": {0: "image_batch_size"},
+            "pooler_output": {0: "image_batch_size"},
+        }
 @register_in_tasks_manager(
     "vision-encoder-decoder",
     *[

tests/openvino/test_exporters_cli.py

@@ -116,6 +116,13 @@ class OVCLIExportTestCase(unittest.TestCase):
                 ("text-to-video", "ltx-video"),
             ]
         )
+
+    if is_transformers_version(">=", "4.56"):
+        SUPPORTED_ARCHITECTURES.extend(
+            [
+                ("feature-extraction", "vit-with-pooler"),
+            ]
+        )
 
     EXPECTED_NUMBER_OF_TOKENIZER_MODELS = {
         "gpt2": 2 if is_tokenizers_version("<", "0.20") or is_openvino_version(">=", "2024.5") else 0,

tests/openvino/test_modeling.py

@@ -1573,6 +1573,7 @@ def test_compare_with_and_without_past_key_values(self):
 class OVModelForCustomTasksIntegrationTest(unittest.TestCase):
     SUPPORTED_ARCHITECTURES_WITH_ATTENTION = ["vit-with-attentions"]
     SUPPORTED_ARCHITECTURES_WITH_HIDDEN_STATES = ["vit-with-hidden-states"]
+    SUPPORTED_ARCHITECTURES_WITH_HIDDEN_STATES = ["dinov3_vit"]
 
     def _get_sample_image(self):
         url = TEST_IMAGE_URL

tests/openvino/utils_tests.py

@@ -199,6 +199,7 @@
     "sana": "katuni4ka/tiny-random-sana",
     "sana-sprint": "katuni4ka/tiny-random-sana-sprint",
     "ltx-video": "katuni4ka/tiny-random-ltx-video",
+    "dinov3_vit": "snake7gun/tiny-random-dinov3",
 }