Ultralytics YOLOUnclaimed

Ultralytics YOLO is a family of computer vision models for analysing images and video in real time. The site presents the models as a way to power vision AI solutions, covering efficient tasks such as object detection, instance segmentation, pose estimation, classification and oriented bounding box detection. The current model is YOLO26, and the page says the nano variant delivers up to 43% faster CPU inference.

The models are aimed at teams building vision features into products and operations. The site lists use across industries including manufacturing, transportation and logistics, retail, healthcare, food and beverage, and environmental monitoring. Its customer examples describe work such as redacting property videos, detecting defects on conveyors, monitoring parking and inspecting power lines.

The page says the models can be deployed on edge devices, in cloud infrastructure and in hybrid systems. Developers can try YOLO26 inference in the browser by uploading an image, adjusting confidence, IoU and image size, and viewing predictions. For a dedicated endpoint, the site gives a Python example that posts an image to a deployed URL with an API key.

Licensing is split into paths. The open-source framework is available under AGPL 3.0, and the site also describes academic and enterprise licences for commercial use. The Platform page lists a Free plan at $0 per month with 100 GB of storage, 100 models and 3 cloud deployments. A Pro plan is $29 per seat per month, with 500 GB of storage, 500 models, 10 cloud deployments and up to five team members. Enterprise pricing is available on request and adds SSO, HIPAA BAA for eligible customers, and on-premise options.

The Platform includes manual annotation across detection, segmentation, classification, pose and oriented bounding boxes, plus SAM-based and YOLO-based smart annotation. Cloud GPUs are listed from $0.24 per hour. The company states it holds ISO/IEC 27001:2022 certification and SOC 2 Type I attestation.