diff --git a/global.go b/global.go
new file mode 100644
index 0000000..bfbc2c9
--- /dev/null
+++ b/global.go
@@ -0,0 +1,16 @@
+package main
+
+import ort "github.com/yalue/onnxruntime_go"
+
+var (
+	UseCoreML  = false
+	Blank      []float32
+	ModelPath  = "./yolov8m.onnx"
+	Yolo8Model ModelSession
+)
+
+type ModelSession struct {
+	Session *ort.AdvancedSession
+	Input   *ort.Tensor[float32]
+	Output  *ort.Tensor[float32]
+}
diff --git a/go.mod b/go.mod
index 4037770..b97d654 100644
--- a/go.mod
+++ b/go.mod
@@ -2,6 +2,7 @@ module object_detector
 
 go 1.18
 
-require github.com/nfnt/resize v0.0.0-20180221191011-83c6a9932646
-
-require github.com/yalue/onnxruntime_go v0.0.0-20230331205425-1acf4f2a2e42 // indirect
+require (
+	github.com/nfnt/resize v0.0.0-20180221191011-83c6a9932646
+	github.com/yalue/onnxruntime_go v1.3.0
+)
diff --git a/imagex.go b/imagex.go
new file mode 100644
index 0000000..fd68456
--- /dev/null
+++ b/imagex.go
@@ -0,0 +1,109 @@
+package main
+
+import (
+	"github.com/nfnt/resize"
+	"image"
+	"io"
+	"math"
+	"sort"
+)
+
+// Function used to convert RAW output from YOLOv8 to an array
+// of detected objects. Each object contain the bounding box of
+// this object, the type of object and the probability
+// Returns array of detected objects in a format [[x1,y1,x2,y2,object_type,probability],..]
+func process_output(output []float32, img_width, img_height int64) [][]interface{} {
+	boxes := [][]interface{}{}
+	for index := 0; index < 8400; index++ {
+		class_id, prob := 0, float32(0.0)
+		for col := 0; col < 80; col++ {
+			if output[8400*(col+4)+index] > prob {
+				prob = output[8400*(col+4)+index]
+				class_id = col
+			}
+		}
+		if prob < 0.5 {
+			continue
+		}
+		label := yolo_classes[class_id]
+		xc := output[index]
+		yc := output[8400+index]
+		w := output[2*8400+index]
+		h := output[3*8400+index]
+		x1 := (xc - w/2) / 640 * float32(img_width)
+		y1 := (yc - h/2) / 640 * float32(img_height)
+		x2 := (xc + w/2) / 640 * float32(img_width)
+		y2 := (yc + h/2) / 640 * float32(img_height)
+		boxes = append(boxes, []interface{}{float64(x1), float64(y1), float64(x2), float64(y2), label, prob})
+	}
+
+	sort.Slice(boxes, func(i, j int) bool {
+		return boxes[i][5].(float32) < boxes[j][5].(float32)
+	})
+	result := [][]interface{}{}
+	for len(boxes) > 0 {
+		result = append(result, boxes[0])
+		tmp := [][]interface{}{}
+		for _, box := range boxes {
+			if iou(boxes[0], box) < 0.7 {
+				tmp = append(tmp, box)
+			}
+		}
+		boxes = tmp
+	}
+	return result
+}
+
+// Function calculates "Intersection-over-union" coefficient for specified two boxes
+// https://pyimagesearch.com/2016/11/07/intersection-over-union-iou-for-object-detection/.
+// Returns Intersection over union ratio as a float number
+func iou(box1, box2 []interface{}) float64 {
+	return intersection(box1, box2) / union(box1, box2)
+}
+
+// Function calculates union area of two boxes
+// Returns Area of the boxes union as a float number
+func union(box1, box2 []interface{}) float64 {
+	box1_x1, box1_y1, box1_x2, box1_y2 := box1[0].(float64), box1[1].(float64), box1[2].(float64), box1[3].(float64)
+	box2_x1, box2_y1, box2_x2, box2_y2 := box2[0].(float64), box2[1].(float64), box2[2].(float64), box2[3].(float64)
+	box1_area := (box1_x2 - box1_x1) * (box1_y2 - box1_y1)
+	box2_area := (box2_x2 - box2_x1) * (box2_y2 - box2_y1)
+	return box1_area + box2_area - intersection(box1, box2)
+}
+
+// Function calculates intersection area of two boxes
+// Returns Area of intersection of the boxes as a float number
+func intersection(box1, box2 []interface{}) float64 {
+	box1_x1, box1_y1, box1_x2, box1_y2 := box1[0].(float64), box1[1].(float64), box1[2].(float64), box1[3].(float64)
+	box2_x1, box2_y1, box2_x2, box2_y2 := box2[0].(float64), box2[1].(float64), box2[2].(float64), box2[3].(float64)
+	x1 := math.Max(box1_x1, box2_x1)
+	y1 := math.Max(box1_y1, box2_y1)
+	x2 := math.Min(box1_x2, box2_x2)
+	y2 := math.Min(box1_y2, box2_y2)
+	return (x2 - x1) * (y2 - y1)
+}
+
+// Function used to convert input image to tensor,
+// required as an input to YOLOv8 object detection
+// network.
+// Returns the input tensor, original image width and height
+func prepare_input(buf io.Reader) ([]float32, int64, int64) {
+	img, _, _ := image.Decode(buf)
+	size := img.Bounds().Size()
+	img_width, img_height := int64(size.X), int64(size.Y)
+	img = resize.Resize(640, 640, img, resize.Lanczos3)
+	red := []float32{}
+	green := []float32{}
+	blue := []float32{}
+	for y := 0; y < 640; y++ {
+		for x := 0; x < 640; x++ {
+			r, g, b, _ := img.At(x, y).RGBA()
+			red = append(red, float32(r/257)/255.0)
+			green = append(green, float32(g/257)/255.0)
+			blue = append(blue, float32(b/257)/255.0)
+		}
+	}
+	input := append(red, green...)
+	input = append(input, blue...)
+	return input, img_width, img_height
+}
diff --git a/main.go b/main.go
index 1f88a9c..2c3b9c5 100644
--- a/main.go
+++ b/main.go
@@ -2,17 +2,13 @@ package main
 
 import (
 	"encoding/json"
-	"github.com/nfnt/resize"
-	ort "github.com/yalue/onnxruntime_go"
-	"image"
+	"fmt"
 	_ "image/gif"
 	_ "image/jpeg"
 	_ "image/png"
 	"io"
-	"math"
 	"net/http"
 	"os"
-	"sort"
 )
 
 // Main function that defines
@@ -43,7 +39,10 @@ func index(w http.ResponseWriter, _ *http.Request) {
 func detect(w http.ResponseWriter, r *http.Request) {
 	r.ParseMultipartForm(0)
 	file, _, _ := r.FormFile("image_file")
-	boxes := detect_objects_on_image(file)
+	boxes, err := detect_objects_on_image(file)
+	if err != nil {
+		fmt.Println(err.Error())
+	}
 	buf, _ := json.Marshal(&boxes)
 	w.Write(buf)
 }
@@ -53,143 +52,33 @@ func detect(w http.ResponseWriter, r *http.Request) {
 // and returns an array of detected objects
 // and their bounding boxes
 // Returns Array of bounding boxes in format [[x1,y1,x2,y2,object_type,probability],..]
-func detect_objects_on_image(buf io.Reader) [][]interface{} {
+func detect_objects_on_image(buf io.Reader) ([][]interface{}, error) {
 	input, img_width, img_height := prepare_input(buf)
-	output := run_model(input)
-	return process_output(output, img_width, img_height)
-}
-
-// Function used to convert input image to tensor,
-// required as an input to YOLOv8 object detection
-// network.
-// Returns the input tensor, original image width and height
-func prepare_input(buf io.Reader) ([]float32, int64, int64) {
-	img, _, _ := image.Decode(buf)
-	size := img.Bounds().Size()
-	img_width, img_height := int64(size.X), int64(size.Y)
-	img = resize.Resize(640, 640, img, resize.Lanczos3)
-	red := []float32{}
-	green := []float32{}
-	blue := []float32{}
-	for y := 0; y < 640; y++ {
-		for x := 0; x < 640; x++ {
-			r, g, b, _ := img.At(x, y).RGBA()
-			red = append(red, float32(r/257)/255.0)
-			green = append(green, float32(g/257)/255.0)
-			blue = append(blue, float32(b/257)/255.0)
-		}
+	output, err := run_model(input)
+	if err != nil {
+		return nil, err
 	}
-	input := append(red, green...)
-	input = append(input, blue...)
-	return input, img_width, img_height
+
+	data := process_output(output, img_width, img_height)
+
+	return data, nil
 }
 
 // Function used to pass provided input tensor to
 // YOLOv8 neural network and return result
 // Returns raw output of YOLOv8 network as a single dimension
 // array
-func run_model(input []float32) []float32 {
-	ort.SetSharedLibraryPath("./libonnxruntime.so")
-	_ = ort.InitializeEnvironment()
+func run_model(input []float32) ([]float32, error) {
 
-	inputShape := ort.NewShape(1, 3, 640, 640)
-	inputTensor, _ := ort.NewTensor(inputShape, input)
+	var err error
 
-	outputShape := ort.NewShape(1, 84, 8400)
-	outputTensor, _ := ort.NewEmptyTensor[float32](outputShape)
-
-	session, _ := ort.NewSession[float32]("./yolov8m.onnx",
-		[]string{"images"}, []string{"output0"},
-		[]*ort.Tensor[float32]{inputTensor}, []*ort.Tensor[float32]{outputTensor})
-
-	_ = session.Run()
-	return outputTensor.GetData()
-}
-
-// Function used to convert RAW output from YOLOv8 to an array
-// of detected objects. Each object contain the bounding box of
-// this object, the type of object and the probability
-// Returns array of detected objects in a format [[x1,y1,x2,y2,object_type,probability],..]
-func process_output(output []float32, img_width, img_height int64) [][]interface{} {
-	boxes := [][]interface{}{}
-	for index := 0; index < 8400; index++ {
-		class_id, prob := 0, float32(0.0)
-		for col := 0; col < 80; col++ {
-			if output[8400*(col+4)+index] > prob {
-				prob = output[8400*(col+4)+index]
-				class_id = col
-			}
+	if Yolo8Model.Session == nil {
+		Yolo8Model, err = InitYolo8Session(input)
+		if err != nil {
+			return nil, err
 		}
-		if prob < 0.5 {
-			continue
-		}
-		label := yolo_classes[class_id]
-		xc := output[index]
-		yc := output[8400+index]
-		w := output[2*8400+index]
-		h := output[3*8400+index]
-		x1 := (xc - w/2) / 640 * float32(img_width)
-		y1 := (yc - h/2) / 640 * float32(img_height)
-		x2 := (xc + w/2) / 640 * float32(img_width)
-		y2 := (yc + h/2) / 640 * float32(img_height)
-		boxes = append(boxes, []interface{}{float64(x1), float64(y1), float64(x2), float64(y2), label, prob})
 	}
 
-	sort.Slice(boxes, func(i, j int) bool {
-		return boxes[i][5].(float32) < boxes[j][5].(float32)
-	})
-	result := [][]interface{}{}
-	for len(boxes) > 0 {
-		result = append(result, boxes[0])
-		tmp := [][]interface{}{}
-		for _, box := range boxes {
-			if iou(boxes[0], box) < 0.7 {
-				tmp = append(tmp, box)
-			}
-		}
-		boxes = tmp
-	}
-	return result
-}
+	return runInference(Yolo8Model, input)
 
-// Function calculates "Intersection-over-union" coefficient for specified two boxes
-// https://pyimagesearch.com/2016/11/07/intersection-over-union-iou-for-object-detection/.
-// Returns Intersection over union ratio as a float number
-func iou(box1, box2 []interface{}) float64 {
-	return intersection(box1, box2) / union(box1, box2)
-}
-
-// Function calculates union area of two boxes
-// Returns Area of the boxes union as a float number
-func union(box1, box2 []interface{}) float64 {
-	box1_x1, box1_y1, box1_x2, box1_y2 := box1[0].(float64), box1[1].(float64), box1[2].(float64), box1[3].(float64)
-	box2_x1, box2_y1, box2_x2, box2_y2 := box2[0].(float64), box2[1].(float64), box2[2].(float64), box2[3].(float64)
-	box1_area := (box1_x2 - box1_x1) * (box1_y2 - box1_y1)
-	box2_area := (box2_x2 - box2_x1) * (box2_y2 - box2_y1)
-	return box1_area + box2_area - intersection(box1, box2)
-}
-
-// Function calculates intersection area of two boxes
-// Returns Area of intersection of the boxes as a float number
-func intersection(box1, box2 []interface{}) float64 {
-	box1_x1, box1_y1, box1_x2, box1_y2 := box1[0].(float64), box1[1].(float64), box1[2].(float64), box1[3].(float64)
-	box2_x1, box2_y1, box2_x2, box2_y2 := box2[0].(float64), box2[1].(float64), box2[2].(float64), box2[3].(float64)
-	x1 := math.Max(box1_x1, box2_x1)
-	y1 := math.Max(box1_y1, box2_y1)
-	x2 := math.Min(box1_x2, box2_x2)
-	y2 := math.Min(box1_y2, box2_y2)
-	return (x2 - x1) * (y2 - y1)
-}
-
-// Array of YOLOv8 class labels
-var yolo_classes = []string{
-	"person", "bicycle", "car", "motorcycle", "airplane", "bus", "train", "truck", "boat",
-	"traffic light", "fire hydrant", "stop sign", "parking meter", "bench", "bird", "cat", "dog", "horse",
-	"sheep", "cow", "elephant", "bear", "zebra", "giraffe", "backpack", "umbrella", "handbag", "tie",
-	"suitcase", "frisbee", "skis", "snowboard", "sports ball", "kite", "baseball bat", "baseball glove",
-	"skateboard", "surfboard", "tennis racket", "bottle", "wine glass", "cup", "fork", "knife", "spoon",
-	"bowl", "banana", "apple", "sandwich", "orange", "broccoli", "carrot", "hot dog", "pizza", "donut",
-	"cake", "chair", "couch", "potted plant", "bed", "dining table", "toilet", "tv", "laptop", "mouse",
-	"remote", "keyboard", "cell phone", "microwave", "oven", "toaster", "sink", "refrigerator", "book",
-	"clock", "vase", "scissors", "teddy bear", "hair drier", "toothbrush",
 }
diff --git a/onnx_util.go b/onnx_util.go
new file mode 100644
index 0000000..1912abc
--- /dev/null
+++ b/onnx_util.go
@@ -0,0 +1,99 @@
+package main
+
+import (
+	ort "github.com/yalue/onnxruntime_go"
+	"runtime"
+)
+
+func InitYolo8Session(input []float32) (ModelSession, error) {
+	ort.SetSharedLibraryPath(getSharedLibPath())
+	err := ort.InitializeEnvironment()
+	if err != nil {
+		return ModelSession{}, err
+	}
+
+	inputShape := ort.NewShape(1, 3, 640, 640)
+	inputTensor, err := ort.NewTensor(inputShape, input)
+	if err != nil {
+		return ModelSession{}, err
+	}
+
+	outputShape := ort.NewShape(1, 84, 8400)
+	outputTensor, err := ort.NewEmptyTensor[float32](outputShape)
+	if err != nil {
+		return ModelSession{}, err
+	}
+
+	options, e := ort.NewSessionOptions()
+	if e != nil {
+		return ModelSession{}, err
+	}
+
+	if UseCoreML { // If CoreML is enabled, append the CoreML execution provider
+		e = options.AppendExecutionProviderCoreML(0)
+		if e != nil {
+			options.Destroy()
+			return ModelSession{}, err
+		}
+		defer options.Destroy()
+	}
+
+	session, err := ort.NewAdvancedSession(ModelPath,
+		[]string{"images"}, []string{"output0"},
+		[]ort.ArbitraryTensor{inputTensor}, []ort.ArbitraryTensor{outputTensor}, options)
+
+	if err != nil {
+		return ModelSession{}, err
+	}
+
+	modelSes := ModelSession{
+		Session: session,
+		Input:   inputTensor,
+		Output:  outputTensor,
+	}
+
+	return modelSes, err
+}
+
+func getSharedLibPath() string {
+	if runtime.GOOS == "windows" {
+		if runtime.GOARCH == "amd64" {
+			return "./third_party/onnxruntime.dll"
+		}
+	}
+	if runtime.GOOS == "darwin" {
+		if runtime.GOARCH == "arm64" {
+			return "./third_party/onnxruntime_arm64.dylib"
+		}
+	}
+	if runtime.GOOS == "linux" {
+		if runtime.GOARCH == "arm64" {
+			return "../third_party/onnxruntime_arm64.so"
+		}
+		return "./third_party/onnxruntime.so"
+	}
+	panic("Unable to find a version of the onnxruntime library supporting this system.")
+}
+
+// Array of YOLOv8 class labels
+var yolo_classes = []string{
+	"person", "bicycle", "car", "motorcycle", "airplane", "bus", "train", "truck", "boat",
+	"traffic light", "fire hydrant", "stop sign", "parking meter", "bench", "bird", "cat", "dog", "horse",
+	"sheep", "cow", "elephant", "bear", "zebra", "giraffe", "backpack", "umbrella", "handbag", "tie",
+	"suitcase", "frisbee", "skis", "snowboard", "sports ball", "kite", "baseball bat", "baseball glove",
+	"skateboard", "surfboard", "tennis racket", "bottle", "wine glass", "cup", "fork", "knife", "spoon",
+	"bowl", "banana", "apple", "sandwich", "orange", "broccoli", "carrot", "hot dog", "pizza", "donut",
+	"cake", "chair", "couch", "potted plant", "bed", "dining table", "toilet", "tv", "laptop", "mouse",
+	"remote", "keyboard", "cell phone", "microwave", "oven", "toaster", "sink", "refrigerator", "book",
+	"clock", "vase", "scissors", "teddy bear", "hair drier", "toothbrush",
+}
+
+func runInference(modelSes ModelSession, input []float32) ([]float32, error) {
+	inTensor := modelSes.Input.GetData()
+	copy(inTensor, input)
+	err := modelSes.Session.Run()
+	if err != nil {
+		return nil, err
+	}
+	return modelSes.Output.GetData(), nil
+}
diff --git a/third_party/onnxruntime.dll b/third_party/onnxruntime.dll
new file mode 100644
index 0000000..872c1a5
Binary files /dev/null and b/third_party/onnxruntime.dll differ
diff --git a/libonnxruntime.so b/third_party/onnxruntime.so
old mode 100755
new mode 100644
similarity index 66%
rename from libonnxruntime.so
rename to third_party/onnxruntime.so
index eb35d62..fa16542
Binary files a/libonnxruntime.so and b/third_party/onnxruntime.so differ
diff --git a/third_party/onnxruntime_arm64.dylib b/third_party/onnxruntime_arm64.dylib
new file mode 100644
index 0000000..13721a7
Binary files /dev/null and b/third_party/onnxruntime_arm64.dylib differ
diff --git a/libonnxruntime.so.1.14.1 b/third_party/onnxruntime_arm64.so
old mode 100755
new mode 100644
similarity index 60%
rename from libonnxruntime.so.1.14.1
rename to third_party/onnxruntime_arm64.so
index eb35d62..80bfcce
Binary files a/libonnxruntime.so.1.14.1 and b/third_party/onnxruntime_arm64.so differ