Formats
CreateML JSON
Object Detection
IMPORTEXPORT
FEATURED
Object Detection

CreateML JSON

Apple's CreateML and Turi Create tools need a special JSON format for object detection tasks.

Overview

When Apple released CreateML in 2018 it was a groundbreaking innovation. For the first time, developers could easily train a machine learning model with no code. No code, that is, except for the one-off scripts they needed to write to munge their data into the proper JSON format that was adopted from Apple's acquired machine learning framework, Turi Create.

Fortunately, Roboflow can both read and write CreateML JSON files so you can truly train a model and export it for use with an Apple iOS device via CoreML without writing a single line of code. Follow our CreateML tutorial to learn how to use it to train a no-code object detection model.

Format Description

Below, learn the structure of CreateML JSON.

createml.json
[
    {
        "image": "0001.jpg",
        "annotations": [
            {
                "label": "helmet",
                "coordinates": {
                    "x": 162.5,
                    "y": 45,
                    "width": 79,
                    "height": 88
                }
            },
            {
                "label": "person",
                "coordinates": {
                    "x": 145.5,
                    "y": 176,
                    "width": 251,
                    "height": 350
                }
            }
        ]
    }
]

Split and Merge Datasets

With Roboflow supervision, an open source Python package with utilities for completing computer vision tasks, you can merge and split detections in CreateML JSON. Read our dedicated guides to learn how to merge and split CreateML JSON detections.

Split detections in CreateML JSON

No items found.

Merge detections in CreateML JSON

No items found.

Supported Models

Below, see model architectures that require data in the CreateML JSON format when training a new model.

On each page below, you can find links to our guides that show how to plot predictions from the model, and complete other common tasks like detecting small objects with the model.

No items found.

On this page

Convert Data from CreateML JSON

Convert to...

Convert Data to CreateML JSON

Convert from...
Formats
CreateML JSON

What is CreateML JSON?

When Apple released CreateML in 2018 it was a groundbreaking innovation. For the first time, developers could easily train a machine learning model with no code. No code, that is, except for the one-off scripts they needed to write to munge their data into the proper JSON format that was adopted from Apple's acquired machine learning framework, Turi Create.

Fortunately, Roboflow can both read and write CreateML JSON files so you can truly train a model and export it for use with an Apple iOS device via CoreML without writing a single line of code. Follow our CreateML tutorial to learn how to use it to train a no-code object detection model.

Deploy Models with Ease on Roboflow

With Roboflow, you can deploy a computer vision model without having to build your own infrastructure.

Get Started Today

CreateML JSON

Below, we show how to convert data to and from

CreateML JSON

. We also list popular models that use the

CreateML JSON

data format. Our conversion tools are free to use.

CONVERT To
CONVERT From
EXAMPLE

Convert Data to CreateML JSON

Use Roboflow to convert
CreateML JSON
to the following formats.

Roboflow is a trusted solution for converting and managing your data. Today, over 250,000 datasets are managed on Roboflow, comprised of 100 million labeled and annotated images.

With Roboflow, you get a solution with:
‍

Free data conversion

SOC II Type 2 Compliant

Trusted by 250,000+ developers

CreateML JSON
 to 
YOLOv8 PyTorch TXT
CreateML JSON
 to 
YOLOv8 Oriented Bounding Boxes
CreateML JSON
 to 
PaliGemma JSONL
CreateML JSON
 to 
YOLOv10 PyTorch TXT
CreateML JSON
 to 
YOLOv9 PyTorch TXT
CreateML JSON
 to 
Florence-2
CreateML JSON
 to 
YOLOv11 PyTorch TXT
CreateML JSON
 to 
OpenAI GPT-4o Object Detection JSONL
CreateML JSON
 to 
COCO Run-Length Encoding (RLE)
CreateML JSON
 to 
COCO JSON
CreateML JSON
 to 
Pascal VOC XML
CreateML JSON
 to 
Tensorflow TFRecord
CreateML JSON
 to 
YOLO Darknet TXT
CreateML JSON
 to 
YOLOv7 PyTorch TXT
CreateML JSON
 to 
YOLOv5 PyTorch TXT
CreateML JSON
 to 
Sagemaker GroundTruth Manifest
CreateML JSON
 to 
Google Cloud AutoML Vision CSV
CreateML JSON
 to 
meituan/yolov6
CreateML JSON
 to 
Scaled-YOLOv4 TXT
CreateML JSON
 to 
YOLO Keras TXT
CreateML JSON
 to 
OpenAI CLIP Classification
CreateML JSON
 to 
RetinaNet Keras CSV
CreateML JSON
 to 
Tensorflow Object Detection CSV
CreateML JSON
 to 
Multiclass Classification CSV
CreateML JSON
 to 
YOLOv4 PyTorch TXT
CreateML JSON
 to 
YOLOv5 Oriented Bounding Boxes

Convert Data from CreateML JSON

Use Roboflow to convert the following formats to
CreateML JSON
format.

Roboflow is a trusted solution for converting and managing your data. Today, over 250,000 datasets are managed on Roboflow, comprised of 100 million labeled and annotated images.

With Roboflow, you get:
‍

Free data conversion

SOC II Type 1 Compliant

Trusted by 250,000+ developers

YOLOv8 PyTorch TXT
to
CreateML JSON
COCO JSON
to
CreateML JSON
Pascal VOC XML
to
CreateML JSON
YOLO Darknet TXT
to
CreateML JSON
Sagemaker GroundTruth Manifest
to
CreateML JSON
Google Cloud AutoML Vision CSV
to
CreateML JSON
LabelMe JSON
to
CreateML JSON
SuperAnnotate JSON
to
CreateML JSON
Supervisely JSON
to
CreateML JSON
Scale AI JSON
to
CreateML JSON
LabelBox JSON
to
CreateML JSON
VGG Image Annotator JSON
to
CreateML JSON
VGG Image Annotator CSV
to
CreateML JSON
VoTT CSV
to
CreateML JSON
VoTT JSON
to
CreateML JSON
LabelBox Video JSON
to
CreateML JSON
IBM Cloud Annotations JSON
to
CreateML JSON
Unity Perception JSON
to
CreateML JSON
YOLO Keras TXT
to
CreateML JSON
OIDv4 TXT
to
CreateML JSON
OpenImages CSV
to
CreateML JSON
RetinaNet Keras CSV
to
CreateML JSON
Tensorflow Object Detection CSV
to
CreateML JSON
Multiclass Classification CSV
to
CreateML JSON
Cogniac
to
CreateML JSON
Kaggle Wheat CSV
to
CreateML JSON
Udacity TXT
to
CreateML JSON
Marmot XML
to
CreateML JSON
Roboflow is a trusted solution for converting and managing your data. Today, over 100,000 datasets are managed on Roboflow, comprised of 100 million labeled and annotated images.

Once your data is in Roboflow, just add the link from your dataset and you're ready to go. We even include the code to export to common inference formats like TFLite, ONNX, and CoreML.

Below are pre-configured models that use the
CreateML JSON
data format
.

What computer vision models use CreateML JSON?

The

No items found.

models all use the

CreateML JSON

data format.

We don't currently have models that use this annotation format.
To see our entire list of computer vision models, check out the Roboflow Model Library.
createml.json
[
    {
        "image": "0001.jpg",
        "annotations": [
            {
                "label": "helmet",
                "coordinates": {
                    "x": 162.5,
                    "y": 45,
                    "width": 79,
                    "height": 88
                }
            },
            {
                "label": "person",
                "coordinates": {
                    "x": 145.5,
                    "y": 176,
                    "width": 251,
                    "height": 350
                }
            }
        ]
    }
]
Ready to get started?

Join over 1 million developers building with Roboflow

Try it for FreeTalk to Sales