Our pricing plans fit the needs of enterprises, start-ups, researchers and hobbyists.


Ideal for evaluating and labeling large public datasets

All labeling solutions

All curation solutions

API and Python SDK access

Unlimited images in public datasets

3,000 images in private datasets

Get started

Ideal for labeling private datasets with a team

Everything in  Free  

50,000 images in private datasets

300 hours of labeling per month

5 collaborators per dataset

Extended email support

Contact us



Ideal for labeling large datasets with advanced automation

Everything in  Team  

Custom labeling editors

Tailored solutions

Priority Slack support

Contact us

Labeling-as-a-service solutions

Prefer to outsource your labeling efforts? We've got you covered.

Let us handle the heavy lifting with our labeling expertise.

Send us your data via the API along with the labeling requirements, and we'll provide you with perfect ground truth data. We're computer vision experts with a soft spot for image segmentation. Contact us to get started!

Instance segmentation

$3.90 / Image

Semantic segmentation

$3.50 / Image


$0.07 / Image
+ $0.07 / Annotation

Bounding boxes

$0.07 / Image
+ $0.07 / Annotation


$0.07 / Image
+ $0.07 / Annotation

* Minimum batch sizes apply. Additional commitment and project-specific discounts are available.

Academic program

We’re collaborating with students and researchers worldwide to solve some of the hardest computer vision problems. is 100% free for non-commercial academic use.

Some of our academic users include ...

More questions?

As computer vision engineers, we understand that the best models are trained on the best data. offers a platform to curate your dataset by enabling a quick iteration between data labeling, model training and failure case discovery. Adjust distribution biases, uncover edge cases, correct labels, investigate image-level model performance and slice the dataset, among many other data and model curation features.

More info

We have a soft spot for segmentation, but support other labeling types such as classification and bounding box detection as well. Other labeling types are on the roadmap. You can also embed your own custom labeling interface on - reach out to us for more information.

Reach out

We provide a standard flexible format for each annotation type as described in our documentation. You can use our API and Python SDK to convert this into other formats. Reach out to us for tailored support.

Reach out

Yes, we have access to multiple workforce partners who have experience working with our platform. Reach out to us to get access to an experienced workforce. You can also easily onboard your own labeling workforce and receive a hands-on workshop getting your workforce team up to speed as fast as possible.

Reach out

The data you share with us will always be owned by you. We do not share your data with third parties, except for agreed-upon workforce purposes. Our terms describe this in more detail.

Reach out for more info

We strongly believe in iterative workflows and data-first approaches. You can use our platform to easily select which data to label, fix distribution biases, upload prediction scores and labeling priorities, and so much more.

Request a personal demo

The computer vision platform to build better datasets iteratively

Explore the platform, or contact us for more info.

Explore the platform Request demo