diff --git a/core/variant/variant_utility.cpp b/core/variant/variant_utility.cpp
index c1a0ad73b09..4b38bf5716e 100644
--- a/core/variant/variant_utility.cpp
+++ b/core/variant/variant_utility.cpp
@@ -235,7 +235,44 @@ struct VariantUtilityFunctions {
 		return Math::snapped(value, step);
 	}
 
-	static inline double lerp(double from, double to, double weight) {
+	static inline Variant lerp(const Variant &from, const Variant &to, double weight, Callable::CallError &r_error) {
+		r_error.error = Callable::CallError::CALL_OK;
+		if (from.get_type() != to.get_type()) {
+			r_error.error = Callable::CallError::CALL_ERROR_INVALID_ARGUMENT;
+			r_error.argument = 1;
+			return Variant();
+		}
+
+		switch (from.get_type()) {
+			case Variant::FLOAT: {
+				return lerpf(VariantInternalAccessor<double>::get(&from), to, weight);
+			} break;
+			case Variant::VECTOR2: {
+				return VariantInternalAccessor<Vector2>::get(&from).lerp(VariantInternalAccessor<Vector2>::get(&to), weight);
+			} break;
+			case Variant::VECTOR3: {
+				return VariantInternalAccessor<Vector3>::get(&from).lerp(VariantInternalAccessor<Vector3>::get(&to), weight);
+			} break;
+			case Variant::VECTOR4: {
+				return VariantInternalAccessor<Vector4>::get(&from).lerp(VariantInternalAccessor<Vector4>::get(&to), weight);
+			} break;
+			case Variant::QUATERNION: {
+				return VariantInternalAccessor<Quaternion>::get(&from).slerp(VariantInternalAccessor<Quaternion>::get(&to), weight);
+			} break;
+			case Variant::BASIS: {
+				return VariantInternalAccessor<Basis>::get(&from).slerp(VariantInternalAccessor<Basis>::get(&to), weight);
+			} break;
+			case Variant::COLOR: {
+				return VariantInternalAccessor<Color>::get(&from).lerp(VariantInternalAccessor<Color>::get(&to), weight);
+			} break;
+			default: {
+				r_error.error = Callable::CallError::CALL_ERROR_INVALID_METHOD;
+				return Variant();
+			}
+		}
+	}
+
+	static inline double lerpf(double from, double to, double weight) {
 		return Math::lerp(from, to, weight);
 	}
 
@@ -1262,7 +1299,6 @@ void Variant::_register_variant_utility_functions() {
 	FUNCBINDR(absi, sarray("x"), Variant::UTILITY_FUNC_TYPE_MATH);
 
 	FUNCBINDVR(sign, sarray("x"), Variant::UTILITY_FUNC_TYPE_MATH);
-
 	FUNCBINDR(signf, sarray("x"), Variant::UTILITY_FUNC_TYPE_MATH);
 	FUNCBINDR(signi, sarray("x"), Variant::UTILITY_FUNC_TYPE_MATH);
 
@@ -1280,7 +1316,8 @@ void Variant::_register_variant_utility_functions() {
 	FUNCBINDR(step_decimals, sarray("x"), Variant::UTILITY_FUNC_TYPE_MATH);
 	FUNCBINDR(snapped, sarray("x", "step"), Variant::UTILITY_FUNC_TYPE_MATH);
 
-	FUNCBINDR(lerp, sarray("from", "to", "weight"), Variant::UTILITY_FUNC_TYPE_MATH);
+	FUNCBINDVR3(lerp, sarray("from", "to", "weight"), Variant::UTILITY_FUNC_TYPE_MATH);
+	FUNCBINDR(lerpf, sarray("from", "to", "weight"), Variant::UTILITY_FUNC_TYPE_MATH);
 	FUNCBINDR(cubic_interpolate, sarray("from", "to", "pre", "post", "weight"), Variant::UTILITY_FUNC_TYPE_MATH);
 	FUNCBINDR(bezier_interpolate, sarray("start", "control_1", "control_2", "end", "t"), Variant::UTILITY_FUNC_TYPE_MATH);
 	FUNCBINDR(lerp_angle, sarray("from", "to", "weight"), Variant::UTILITY_FUNC_TYPE_MATH);
diff --git a/doc/classes/@GlobalScope.xml b/doc/classes/@GlobalScope.xml
index 4a16244235a..6c7199c689b 100644
--- a/doc/classes/@GlobalScope.xml
+++ b/doc/classes/@GlobalScope.xml
@@ -439,16 +439,18 @@
 			</description>
 		</method>
 		<method name="lerp">
-			<return type="float" />
-			<argument index="0" name="from" type="float" />
-			<argument index="1" name="to" type="float" />
-			<argument index="2" name="weight" type="float" />
+			<return type="Variant" />
+			<argument index="0" name="from" type="Variant" />
+			<argument index="1" name="to" type="Variant" />
+			<argument index="2" name="weight" type="Variant" />
 			<description>
 				Linearly interpolates between two values by the factor defined in [code]weight[/code]. To perform interpolation, [code]weight[/code] should be between [code]0.0[/code] and [code]1.0[/code] (inclusive). However, values outside this range are allowed and can be used to perform [i]extrapolation[/i]. Use [method clamp] on the result of [method lerp] if this is not desired.
+				Both [code]from[/code] and [code]to[/code] must have matching types. Supported types: [float], [Vector2], [Vector3], [Vector4], [Color], [Quaternion], [Basis].
 				[codeblock]
 				lerp(0, 4, 0.75) # Returns 3.0
 				[/codeblock]
 				See also [method inverse_lerp] which performs the reverse of this operation. To perform eased interpolation with [method lerp], combine it with [method ease] or [method smoothstep]. See also [method range_lerp] to map a continuous series of values to another.
+				[b]Note:[/b] For better type safety, you can use [method lerpf], [method Vector2.lerp], [method Vector3.lerp], [method Vector4.lerp], [method Color.lerp], [method Quaternion.slerp] or [method Basis.slerp] instead.
 			</description>
 		</method>
 		<method name="lerp_angle">
@@ -471,6 +473,19 @@
 				[b]Note:[/b] This method lerps through the shortest path between [code]from[/code] and [code]to[/code]. However, when these two angles are approximately [code]PI + k * TAU[/code] apart for any integer [code]k[/code], it's not obvious which way they lerp due to floating-point precision errors. For example, [code]lerp_angle(0, PI, weight)[/code] lerps counter-clockwise, while [code]lerp_angle(0, PI + 5 * TAU, weight)[/code] lerps clockwise.
 			</description>
 		</method>
+		<method name="lerpf">
+			<return type="float" />
+			<argument index="0" name="from" type="float" />
+			<argument index="1" name="to" type="float" />
+			<argument index="2" name="weight" type="float" />
+			<description>
+				Linearly interpolates between two values by the factor defined in [code]weight[/code]. To perform interpolation, [code]weight[/code] should be between [code]0.0[/code] and [code]1.0[/code] (inclusive). However, values outside this range are allowed and can be used to perform [i]extrapolation[/i].
+				[codeblock]
+				lerp(0, 4, 0.75) # Returns 3.0
+				[/codeblock]
+				See also [method inverse_lerp] which performs the reverse of this operation. To perform eased interpolation with [method lerp], combine it with [method ease] or [method smoothstep].
+			</description>
+		</method>
 		<method name="linear2db">
 			<return type="float" />
 			<argument index="0" name="lin" type="float" />