diff --git a/src/util/vec/float3.h b/src/util/vec/float3.h
new file mode 100644
index 0000000..6a223ad
--- /dev/null
+++ b/src/util/vec/float3.h
@@ -0,0 +1,68 @@
+#pragma once
+
+#include <math.h>
+
+#include "../atrb.h"
+
+// stores a 3D point using a floating-point number
+struct float3 {
+	float x;
+	float y;
+	float z;
+};
+typedef struct float3 float3;
+
+// adds one float3 to another
+atrb_const static inline float3 float3_add(float3 v1, float3 v2) {
+	return (float3){v1.x + v2.x, v1.y + v2.y, v1.z + v2.z};
+}
+
+// subtracts one float3 from another
+atrb_const static inline float3 float3_sub(float3 v1, float3 v2) {
+	return (float3){v1.x - v2.x, v1.y - v2.y, v1.z - v2.z};
+}
+
+// multiplies one float3 by another
+atrb_const static inline float3 float3_mul(float3 v1, float3 v2) {
+	return (float3){v1.x * v2.x, v1.y * v2.y, v1.z * v2.z};
+}
+
+// divides one float3 by another
+atrb_const static inline float3 float3_div(float3 v1, float3 v2) {
+	return (float3){v1.x / v2.x, v1.y / v2.y, v1.z / v2.z};
+}
+
+// preforms a scalar multiplication upon the float3 (multiplies the float3 by some value)
+atrb_const static inline float3 float3_mul_s(float3 v, float n) {
+	return (float3){v.x * n, v.y * n, v.z * n};
+}
+
+// preforms a scalar division upon the float3 (divides the float3 by some value)
+atrb_const static inline float3 float3_div_s(float3 v, float n) {
+	return (float3){v.x / n, v.y / n, v.z / n};
+}
+
+atrb_const static inline float3 float3_neg(float3 v) {
+	return (float3){-v.x, -v.y, -v.z};
+}
+
+// gets the squared magnitude/length of float3
+atrb_const static inline float float3_mag2(float3 v) {
+	return (v.x * v.x) + (v.y * v.y) + (v.z * v.z);
+}
+
+// gets the length of float3 (length)
+atrb_const static inline float float3_mag(float3 v) {
+	return sqrtf(float3_mag2(v));
+}
+
+// normalizes the float3
+atrb_const static inline float3 float3_norm(float3 v) {
+	float s = 1.0F / float3_mag(v); // get the scaling factor
+	return float3_mul_s(v, s);      // scale the vector by the scaling factor (slightly more efficient than dividing)
+}
+
+// gets the dot product of two float3s
+atrb_const static inline float float3_dot(float3 v1, float3 v2) {
+	return v1.x * v2.x + v1.y * v2.y + v1.z * v2.z;
+}