Image and video annotation
Label images and videos with bounding boxes, polygons, semantic masks, keypoints and oriented boxes. Smart annotation and team review support detection, segmentation, classification, pose and OBB workflows.
Ultralytics Platform
Freemium
A computer vision platform for annotating data, training YOLO models, and deploying them across cloud and edge environments.
O que é Ultralytics Platform?
Ultralytics Platform brings computer vision development into one workflow: annotate image and video datasets, train Ultralytics YOLO models, and deploy to cloud endpoints or export for edge, mobile and embedded environments. Annotation covers detection, instance and semantic segmentation, classification, pose estimation and oriented object detection. A no-code interface and Python SDK support the workflow. Platform subscriptions and software licensing should be reviewed separately for the intended project.
O que Ultralytics Platform pode fazer?
Cloud model training
Launch training jobs using configurable cloud GPU types and review live metrics and experiment comparisons. The platform supports Ultralytics YOLOv5, YOLOv8, YOLO11, and YOLO26, with access through a no-code interface or Python SDK.
Dataset and team management
Organize datasets, models and projects in a shared workflow with team review and versioning. Team size, storage integrations and workspace limits depend on the current plan.
Model export
Export trained models in 22 formats, including ONNX, TensorRT, CoreML, LiteRT, NCNN, and others, for use across cloud, mobile, edge, and embedded environments.
Deployment and monitoring
Create configurable inference endpoints in global regions. Monitoring includes request metrics, latency, errors, and logs; dedicated endpoints also provide live prediction statistics.
Remote and on-premise workflows
Train on your own hardware while streaming metrics to the platform. Enterprise also offers on-premise data and compute options.
Casos de uso
“Build and maintain vision datasets”
Annotation teams can label images and videos for detection, segmentation, classification, pose, or rotated-object workflows, then organize the resulting data with associated projects and models.
“Train and compare YOLO experiments”
Machine-learning practitioners can run concurrent cloud training jobs, monitor convergence, and compare experiments when selecting a model for an application.
“Deploy and monitor inference services”
Teams serving predictions can deploy regional endpoints and inspect request, latency, error, and log data.
“Target edge and embedded devices”
Developers can export models to formats such as ONNX, TensorRT, CoreML, and LiteRT for deployment beyond the platform's hosted endpoints.
“Analyze sports footage”
Sports teams and developers can build YOLO workflows for player and ball tracking, performance analysis, and coaching insights.
Perguntas frequentes
Can I use the platform without writing code?
Yes. The platform provides a no-code interface, and a Python SDK is available for users who want programmatic control.
Which computer vision tasks are supported?
Annotation covers detection, instance segmentation, semantic segmentation, classification, pose estimation, and oriented object detection. These are the platform’s six annotation task types.
Where can I deploy or export models?
The platform offers dedicated endpoints across global regions and exports in 22 formats for cloud, mobile, edge, and embedded environments, including ONNX, TensorRT, CoreML, LiteRT, and NCNN.
Are platform subscriptions and software licenses the same?
Platform subscriptions and software licensing are separate considerations. Review the current AGPL-3.0 and Ultralytics Enterprise License terms for the intended project; a paid platform subscription should not be assumed to waive software-license obligations.
Can teams train on their own data storage or hardware?
The platform offers a Python SDK and cloud training workflows, while enterprise options include deployment on your infrastructure. Confirm supported storage connections and remote training arrangements for your plan.
What is free, and what costs extra?
The website offers a free platform plan. Cloud compute, deployment, storage, team features and credits depend on the current plan and usage pricing. Check the plan comparison before budgeting a training or deployment workflow.
Informações rápidas
- Product category
- Computer vision development platform
- Primary workflow
- Annotate data, train models, deploy or export
- Model family
- Ultralytics YOLO
- Deployment
- Regional hosted endpoints or exported models for cloud, mobile, and edge environments
- Export formats
- 22
- Platform plans
- Free, Pro, and Enterprise; software licensing is separate
Análise de tráfego de Ultralytics Platform
Os dados de tráfego são apenas para referência.
- Classificação do domínio
- 77
Alternativas ao Ultralytics Platform
Paperspace
www.paperspace.com
Paperspace is a cloud platform for developing, training, and deploying machine learning applications with managed notebooks, GPU machines, and deployment workflows. It serves ML developers, data scientists, researchers, and teams that need on-demand accelerated computing.