All computer vision
in one platform.
Curate, label and build production models for images, videos, 3D, medical and more.
See the whole platform in one video.
Every great model starts as a well-labeled dataset.
Most computer vision teams stitch together a labeling tool, a data lake, training notebooks and deployment scripts — and lose weeks moving data between them.
Supervisely keeps the whole loop in one place: curate your data, label it with AI assistance, train and evaluate models, then put them back to work pre-labeling the next batch.
One loop from raw data to a deployed model.
-
Connect AWS, GCS or Azure without duplicating data. Search images and labels with advanced filters and natural-language AI Search, and find gaps with embeddings maps and class statistics.
Data management -
Label images, videos, 3D point clouds, DICOM, meshes and audio right in the browser. The Smart Tool with Segment Anything 3, auto-tracking and model pre-labeling do the heavy lifting.
Labeling toolboxes -
Train state-of-the-art or custom models on your datasets in a few clicks — on your own GPUs, with experiments tracked and versioned.
Model training -
Get automated performance reports, compare checkpoints against ground truth and each other, and trace weak spots back to the data that caused them.
Model evaluation -
Serve models for assisted labeling, batch inference or production via API, and close the loop with Live Training: the model keeps learning while your team labels.
Serving and inference
A labeling toolbox for every modality, with AI in the loop.
Best-in-class toolboxes for every modality, accelerated by state-of-the-art models for pixel-accurate labels.
Image labeling
Pixel-accurate masks with smart tools, polygons, brushes and bounding boxes — for anything from photos to large images.
Everything in the UI is also in the API.
Python SDK, REST API and App Engine: automate data pipelines, plug in your own models and build custom labeling UIs that run inside Supervisely.
- Apache 2.0
- Open-source Python SDK
- REST
- Full API reference
- App Engine
- Custom UIs in Python
import supervisely as sly
api = sly.Api.from_env()
# Create a project and upload a batch of images
project = api.project.create(WORKSPACE_ID, "Defect inspection")
dataset = api.dataset.create(project.id, "line-04")
api.image.upload_paths(dataset.id, names, paths)
# Pre-label everything with a model deployed in Supervisely
model = api.nn.connect(task_id=MODEL_TASK_ID)
model.predict(project_id=project.id, upload_mode="append")import supervisely as sly
from supervisely.app.widgets import Button, Card, Container, Text
api = sly.Api.from_env()
button = Button("Count images")
result = Text()
@button.click
def count_images():
project = api.project.get_info_by_id(PROJECT_ID)
result.text = f"{project.items_count} images"
layout = Card("My first app", content=Container([button, result]))
app = sly.Application(layout=layout)# Every action in the UI is available over the API
curl -X POST \
https://app.supervisely.com/public/api/v3/projects.list \
-H "x-api-key: $API_TOKEN" \
-H "Content-Type: application/json" \
-d '{"workspaceId": 42}'# Install the Python SDK
pip install supervisely
# Add Supervisely to your MCP client's config
{
"mcpServers": {
"supervisely": {
"command": "supervisely",
"args": ["mcp"],
"env": {
"SERVER_ADDRESS": "https://app.supervisely.com",
"API_TOKEN": "your-api-token"
}
}
}
}- Pre-label the new Line 04 images with our defect detector, then send them to review.
- Search projects Defect inspection · line-04 · 1,284 new images
- Run model defect-detector · 1,284 images pre-labeled
- Create labeling job Review · Line 04 · 3 reviewers
- Done. All 1,284 images are pre-labeled and waiting in “Review · Line 04” for three reviewers.
AI agents can work in it too.
The Python SDK runs an MCP server on your machine. Connect any MCP client, and an agent works in Supervisely with your API token, so it can see and change only what you can.
- Browse data
- Search projects, datasets, items and annotations, and read their stats.
- Label and review
- Create labeling jobs, assign them and follow review status.
- Run models
- Deploy models and pre-label or predict on your data.
- Run apps
- Launch Ecosystem apps, from import and export to training, as tools.
Run Supervisely on your own infrastructure.
Supervisely Enterprise Edition is built for companies scaling their AI infrastructure — in the cloud or on their own servers.
On your servers or in your cloud.
Run Supervisely on your own hardware or in any cloud, on a single server or on Kubernetes, including AWS EKS. Upgrades take one command and back up first.
Enterprise Edition
Here’s why teams trust us.
“A very high product quality.”
BMW Group uses Supervisely to annotate manufacturing images from production lines in their world-wide plants for enhancing quality inspections using deep learning. Supervisely is integrated into the BMW Group AI Platform in order to empower computer vision based AI use cases.
BMW Group · Manufacturing quality inspection“Immensely grateful for their pivotal role in our work”
Working with Supervisely has significantly enhanced our capability to develop AI models for lung CT scans. What sets Supervisely apart is its exceptional support team who are really responsive and adapt to our unique needs with a range of apps and helper files.
“Helps us to do data scientist work better and faster”
We use Supervisely since 2019. The key advantage of this tool is that Supervisely provides a complete data treatment pipeline. An important advantage is that a Supervisely instance can be deployed autonomously on a Client infrastructure, and distributed on different servers.
“A valuable part of our computer vision operations”
Supervisely is a valuable part of our computer vision operations, enabling our teams to efficiently work with camera and LiDAR data across sensor fusion projects. We use Supervisely for its extensive feature set and its ability to adapt to our changing requirements.
“Extremely satisfied with Supervisely and would highly recommend it”
The review tools have been invaluable in ensuring the quality and accuracy. The Python SDK has also been incredibly helpful in automating and streamlining our workflow. The support team on Slack has been extremely helpful and responsive.
“Supervisely excels at that and much more”
We’ve been exceedingly impressed with the customer support, addition of new features, and the flexibility of the publicly available SDK/API. The Supervisely team has also been fast to respond to support questions, and has shown a lot of openness when given feedback on potential improvements.
Your team next?
Talk to a computer vision engineer about your data, your models and your infrastructure.
Book a demo
