Free to start
Free
FREE
- No credit card required
- Import images and labels
- Label images manually
- 50 credits included
- Cloud storage: 1 GB
- Export labeled datasets
Label Your Images.
The fastest bounding box annotation tool for solo developers. Import images, label manually or with AI, and export training-ready datasets in YOLO or COCO format. You own your data — no open-source restrictions, no forced subscription.
Label anything
Annotate images for any computer vision project
Build custom machine learning models for any application.
Prepare your datasets for training and deployment on iOS and Android.
RoboticsAutomationConsumer GoodsLogisticsIndustrialConstructionTransportationHealthcareSportsEntertainmentManufacturingNatureHospitalityEngineeringMarine Science
Easily annotate anything with AI
AI-assisted image annotation — label any object with one click
cat
Annotate all classes and objects with a single click
AI infers which objects you're labelling directly from the class name. Simply create your classes and then click "Label with AI". Try it now.
AI labeling time and accuracy can vary
Lightning fast bounding box labels
The fastest bounding box labelling tool for object detection
UI optimisation to minimise unnecessary mouse movements.
Infinitely scroll through your dataset.
Keyboard shortcuts to switch classes instantly.
Batch labelling with AI
Batch AI annotation — label your entire dataset in seconds
Label your entire dataset in seconds.
Annotate multiple images and classes with one click.
Customise thresholds for each class to capture tricky subjects.
Refine bounding boxes created by AI to get the perfect fit.
Plans and pricing
Start free, upgrade whenever.
Plans include storage plus credits. Annual plans pre-pay a full year of credits upfront.
Monthly plans renew every month with the same included credit allowance.
Best for getting started
Starter
£8.00/mo
- 500 credits per month
- 5 GB cloud storage
- Monthly or annual billing
- Manage projects and exports in one workspace
Most Popular
Growth
£23.00/mo
- 2,500 credits per month
- 25 GB cloud storage
- Faster throughput for active datasets
- Annual billing pre-pays 30,000 credits per year
High throughput
Scale
£75.00/mo
- 10,000 credits per month
- 100 GB cloud storage
- Designed for large-scale AI labeling
- Annual billing pre-pays 120,000 credits per year
The Story Behind BBoxML
I’m Joel Gibbard, a robotics engineer and entrepreneur. My background in machine vision started during my Robotics degree, where I first learned about the fundamentals of machine learning and image processing. However, for the next decade, my focus shifted towards designing and building robotic hands with Open Bionics. That journey in assistive tech eventually earned me an MBE for services to innovation and engineering.
I returned to machine vision recently while building Sxratch, an app designed to help golfers improve their putting. I expected the tooling to have matured; instead, I hit a wall.
Free tools often had open-source licensing restrictions or required expensive subscriptions. AI-assisted labeling wasn't precise enough for the small, bespoke objects, like golf balls and holes, that my model required.
I didn't want a bloated enterprise subscription; I just wanted to label my data accurately and own the output. When I couldn't find a tool that respected an engineer's time and data, I built my own. BBoxML is the result: a high-speed annotation tool designed for people who need precision, and want to build for commercialisation from the start.
Questions people ask before they start labelling
Answers to the most common questions about image annotation, YOLO and COCO exports, AI labelling, and getting started for free.
What's the difference between YOLO and COCO annotation formats?
YOLO stores one text file per image with normalized bounding box coordinates and class IDs. COCO stores the whole dataset in a JSON file with images, categories, and annotations together. YOLO is usually simpler for training YOLO models directly, while COCO is more flexible for dataset interchange and tooling.
How do I label images for YOLO training?
Create a project, upload your images, create your classes, draw bounding boxes around the objects you want to detect, then export the dataset in YOLO format. Each image should be labeled consistently using the same class names and box rules across the dataset.
Can I use AI to label my images automatically?
Yes. BBoxML includes AI-assisted labelling so you can speed up annotation and then review the results manually. AI labels should still be checked before training to make sure they match your dataset standards.
How many images do I need to train an object detection model?
It depends on the difficulty of the task, the number of classes, and how much variation your model needs to learn. Small experiments may start with a few hundred labeled images, but production-quality models often need far more examples across different backgrounds, angles, and lighting conditions.
Can I annotate images online in my browser?
Yes. BBoxML runs in the browser, so you can upload images, draw bounding boxes, manage classes, and export datasets online without installing desktop annotation software.
Can I import labels I've already created?
Yes. BBoxML supports importing existing labeled datasets so you can continue annotating, review older labels, or prepare exports without starting again from scratch.
What image formats does BBoxML support?
BBoxML supports common image formats used in annotation workflows, including JPG, JPEG, PNG, WebP, TIFF, BMP, GIF, HEIC, and related browser-compatible image types.
Can I annotate images for free?
Yes. The Free plan lets you start annotating images in BBoxML without a credit card. You can label images manually, work in the browser, and use the included free-tier limits shown on the pricing section.
Are there license restrictions on datasets on the free plan?
No special dataset license is added by BBoxML just because you use the Free plan. You are still responsible for making sure you have the rights to use, annotate, train on, and export the datasets you upload.
What is the best online bounding box annotation tool for YOLO datasets?
The best tool depends on your workflow, but many teams want a browser-based tool that supports fast box drawing, class management, AI-assisted labelling, and YOLO export. BBoxML is designed around that workflow so you can label images, review annotations, and export training-ready datasets in one place.
Can I convert COCO annotations to YOLO format?
Yes. If your annotation tool supports both formats, you can usually import an existing COCO dataset and export it as YOLO. BBoxML is designed to help teams work across common object-detection dataset formats without rebuilding annotations from scratch.
Do I need to install software to label images for object detection?
No. With a browser-based annotation tool, you can label images online without installing desktop software. BBoxML is built to let you annotate images, manage classes, and prepare exports directly in your browser.
