diff --git a/neuralnetworks/1.2/types.hal b/neuralnetworks/1.2/types.hal
index 45476eec3d..4ab0c24ed8 100644
--- a/neuralnetworks/1.2/types.hal
+++ b/neuralnetworks/1.2/types.hal
@@ -40,7 +40,7 @@ enum OperandType : @1.0::OperandType {
      * scale is a 32 bit floating point with value greater then zero.
      */
     TENSOR_QUANT16_SYMM = 7,
-    /** A tensor of 16 bit floating point values. */
+    /** A tensor of IEEE 754 16 bit floating point values. */
     TENSOR_FLOAT16 = 8,
     /**
      * A tensor of 8 bit boolean values.
@@ -49,6 +49,8 @@ enum OperandType : @1.0::OperandType {
      * represents false; any other value represents true.
      */
     TENSOR_BOOL8 = 9,
+    /** An IEEE 754 16 bit floating point scalar value. */
+    FLOAT16 = 10,
     /* ADDING A NEW FUNDAMENTAL TYPE REQUIRES UPDATING THE VALUE OF
      * OperandTypeRange::OPERAND_FUNDAMENTAL_MAX.
      */
@@ -62,7 +64,7 @@ enum OperandType : @1.0::OperandType {
  */
 enum OperandTypeRange : uint32_t {
     OPERAND_FUNDAMENTAL_MIN = 0,
-    OPERAND_FUNDAMENTAL_MAX = 9,
+    OPERAND_FUNDAMENTAL_MAX = 10,
     OPERAND_OEM_MIN     = 10000,
     OPERAND_OEM_MAX     = 10001,
 };
diff --git a/neuralnetworks/1.2/vts/functional/ValidateModel.cpp b/neuralnetworks/1.2/vts/functional/ValidateModel.cpp
index 294415bc1b..3fadd480f3 100644
--- a/neuralnetworks/1.2/vts/functional/ValidateModel.cpp
+++ b/neuralnetworks/1.2/vts/functional/ValidateModel.cpp
@@ -152,6 +152,7 @@ static void mutateOperandTypeTest(const sp<IDevice>& device, const Model& model)
 
 static uint32_t getInvalidRank(OperandType type) {
     switch (type) {
+        case OperandType::FLOAT16:
         case OperandType::FLOAT32:
         case OperandType::INT32:
         case OperandType::UINT32:
@@ -183,6 +184,7 @@ static void mutateOperandRankTest(const sp<IDevice>& device, const Model& model)
 
 static float getInvalidScale(OperandType type) {
     switch (type) {
+        case OperandType::FLOAT16:
         case OperandType::FLOAT32:
         case OperandType::INT32:
         case OperandType::UINT32:
@@ -215,6 +217,7 @@ static void mutateOperandScaleTest(const sp<IDevice>& device, const Model& model
 
 static std::vector<int32_t> getInvalidZeroPoints(OperandType type) {
     switch (type) {
+        case OperandType::FLOAT16:
         case OperandType::FLOAT32:
         case OperandType::INT32:
         case OperandType::UINT32:
@@ -258,6 +261,7 @@ static void mutateOperand(Operand* operand, OperandType type) {
     Operand newOperand = *operand;
     newOperand.type = type;
     switch (type) {
+        case OperandType::FLOAT16:
         case OperandType::FLOAT32:
         case OperandType::INT32:
         case OperandType::UINT32: