diff --git a/backends/nvidia/tensorrt/__init__.py b/backends/nvidia/tensorrt/__init__.py
index 73144d28a32..8d6d111f375 100644
--- a/backends/nvidia/tensorrt/__init__.py
+++ b/backends/nvidia/tensorrt/__init__.py
@@ -27,7 +27,9 @@
             sys.path.append(_pkg_dir)
 
 from executorch.backends.nvidia.tensorrt.backend import TensorRTBackend
+from executorch.backends.nvidia.tensorrt.partitioner import TensorRTPartitioner
 
 __all__ = [
     "TensorRTBackend",
+    "TensorRTPartitioner",
 ]
diff --git a/backends/nvidia/tensorrt/partitioner/TARGETS b/backends/nvidia/tensorrt/partitioner/TARGETS
new file mode 100644
index 00000000000..0a42614a385
--- /dev/null
+++ b/backends/nvidia/tensorrt/partitioner/TARGETS
@@ -0,0 +1,5 @@
+load(":targets.bzl", "define_common_targets")
+
+oncall("executorch")
+
+define_common_targets()
diff --git a/backends/nvidia/tensorrt/partitioner/__init__.py b/backends/nvidia/tensorrt/partitioner/__init__.py
new file mode 100644
index 00000000000..b7ac9d4fe4f
--- /dev/null
+++ b/backends/nvidia/tensorrt/partitioner/__init__.py
@@ -0,0 +1,45 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD-style license found in the
+# LICENSE file in the root directory of this source tree.
+
+"""TensorRT partitioner for ExecuTorch."""
+
+from typing import Dict, List, Optional
+
+from executorch.backends.nvidia.tensorrt.backend import TensorRTBackend
+from executorch.exir.backend.compile_spec_schema import CompileSpec
+from executorch.exir.backend.partitioner import (
+    DelegationSpec,
+    Partitioner,
+    PartitionResult,
+)
+from torch.export.exported_program import ExportedProgram
+
+
+class TensorRTPartitioner(Partitioner):
+    """Partitioner for TensorRT backend.
+    """
+
+    def __init__(
+        self,
+        compile_specs: Optional[List[CompileSpec]] = None,
+    ) -> None:
+        super().__init__()
+        self.compile_specs = compile_specs or []
+        self.delegation_spec = DelegationSpec(
+            backend_id=TensorRTBackend.__name__,
+            compile_specs=self.compile_specs,
+        )
+
+    def partition(self, exported_program: ExportedProgram) -> PartitionResult:
+        """Partition the graph for TensorRT delegation.
+
+        Identifies subgraphs that can be lowered to the TensorRT backend.
+        """
+        partition_tags: Dict[str, DelegationSpec] = {}
+        return PartitionResult(
+            tagged_exported_program=exported_program,
+            partition_tags=partition_tags,
+        )
diff --git a/backends/nvidia/tensorrt/partitioner/targets.bzl b/backends/nvidia/tensorrt/partitioner/targets.bzl
new file mode 100644
index 00000000000..4c090b4a767
--- /dev/null
+++ b/backends/nvidia/tensorrt/partitioner/targets.bzl
@@ -0,0 +1,21 @@
+load("@fbsource//xplat/executorch/build:runtime_wrapper.bzl", "runtime")
+
+def define_common_targets():
+    """Defines targets that should be shared between fbcode and xplat.
+
+    The directory containing this targets.bzl file should also contain both
+    TARGETS and BUCK files that call this function.
+    """
+
+    runtime.python_library(
+        name = "partitioner",
+        srcs = [
+            "__init__.py",
+        ],
+        visibility = ["PUBLIC"],
+        deps = [
+            "//caffe2:torch",
+            "//executorch/backends/nvidia/tensorrt:backend",
+            "//executorch/exir/backend:partitioner",
+        ],
+    )