Choosing an Ultralytics model

Default recommendation: YOLO26, pretrained. Latest generation, NMS-free end-to-end (fastest CPU inference, simplest deployment). Use YOLO11/YOLOv8 only to match an existing codebase or a deployment target that doesn't support YOLO26 yet. Most official weights auto-download on first use; sam3.pt requires manual access and download.

Choose in Platform

For the quickest no-code start, open Platform Explore, select Projects, clone the official @ultralytics project for the model family, then train one of its pretrained models on your dataset. The New Model dialog filters base models to the selected dataset task and offers official models plus your own completed checkpoints for further fine-tuning.

Use a Platform model page to inspect metrics, test it in Predict, export it, deploy it, clone it into another project, or download its .pt weights for the Python/CLI workflows below. See Platform Models and Explore.