diff --git a/tensorflow/lite/tools/benchmark/README.md b/tensorflow/lite/tools/benchmark/README.md
index 7ea236120f4..70728f41db1 100644
--- a/tensorflow/lite/tools/benchmark/README.md
+++ b/tensorflow/lite/tools/benchmark/README.md
@@ -69,8 +69,14 @@ and the following optional parameters:
     benchmark tool will not correctly use NNAPI.
 *   `max_delegated_partitions`: `int` (default=0, i.e. no limit) \
     The maximum number of partitions that will be delegated. \
-    Currently supported only by the NNAPI Delegate and it won't work \
-    if `use_legacy_nnapi` has been selected.
+    Currently supported by the Hexagon delegate or the NNAPI delegate but won't
+    work if `use_legacy_nnapi` has been selected.
+*   `min_nodes_per_partition`: `int` (default=0, i.e. default choice implemented
+    by each delegate) \
+    The minimal number of TFLite graph nodes of a partition that needs to be
+    reached to be delegated. A negative value or 0 means to use the default
+    choice of each delegate. \
+    This option is currently only supported by the Hexagon delegate.
 *   `disable_nnapi_cpu`: `bool` (default=false) \
     Excludes the
     [NNAPI CPU reference implementation](https://developer.android.com/ndk/guides/neuralnetworks#device-assignment)
diff --git a/tensorflow/lite/tools/benchmark/default_execution_provider.cc b/tensorflow/lite/tools/benchmark/default_execution_provider.cc
index f7204cba954..1dd8aebccca 100644
--- a/tensorflow/lite/tools/benchmark/default_execution_provider.cc
+++ b/tensorflow/lite/tools/benchmark/default_execution_provider.cc
@@ -28,6 +28,8 @@ class DefaultExecutionProvider : public DelegateProvider {
     default_params_.AddParam("num_threads", BenchmarkParam::Create<int32_t>(1));
     default_params_.AddParam("max_delegated_partitions",
                              BenchmarkParam::Create<int32_t>(0));
+    default_params_.AddParam("min_nodes_per_partition",
+                             BenchmarkParam::Create<int32_t>(0));
   }
 
   std::vector<Flag> CreateFlags(BenchmarkParams* params) const final;
@@ -44,7 +46,12 @@ std::vector<Flag> DefaultExecutionProvider::CreateFlags(
       CreateFlag<int32_t>("num_threads", params,
                           "number of threads used for inference on CPU."),
       CreateFlag<int32_t>("max_delegated_partitions", params,
-                          "Max number of partitions to be delegated.")};
+                          "Max number of partitions to be delegated."),
+      CreateFlag<int32_t>(
+          "min_nodes_per_partition", params,
+          "The minimal number of TFLite graph nodes of a partition that has to "
+          "be reached for it to be delegated.A negative value or 0 means to "
+          "use the default choice of each delegate.")};
   return flags;
 }
 
@@ -53,6 +60,8 @@ void DefaultExecutionProvider::LogParams(const BenchmarkParams& params) const {
                    << params.Get<int32_t>("num_threads") << "]";
   TFLITE_LOG(INFO) << "Max number of delegated partitions : ["
                    << params.Get<int32_t>("max_delegated_partitions") << "]";
+  TFLITE_LOG(INFO) << "Min nodes per partition : ["
+                   << params.Get<int32_t>("min_nodes_per_partition") << "]";
 }
 
 TfLiteDelegatePtr DefaultExecutionProvider::CreateTfLiteDelegate(
diff --git a/tensorflow/lite/tools/benchmark/hexagon_delegate_provider.cc b/tensorflow/lite/tools/benchmark/hexagon_delegate_provider.cc
index 465c71c0a91..1904dd5eebb 100644
--- a/tensorflow/lite/tools/benchmark/hexagon_delegate_provider.cc
+++ b/tensorflow/lite/tools/benchmark/hexagon_delegate_provider.cc
@@ -90,6 +90,8 @@ TfLiteDelegatePtr HexagonDelegateProvider::CreateTfLiteDelegate(
     options.print_graph_profile = params.Get<bool>("hexagon_profiling");
     options.max_delegated_partitions =
         params.Get<int>("max_delegated_partitions");
+    options.min_nodes_per_partition =
+        params.Get<int>("min_nodes_per_partition");
     delegate = evaluation::CreateHexagonDelegate(
         &options, params.Get<std::string>("hexagon_lib_path"));