YOLO26

YOLO26

Finds objects in images and video with the Ultralytics YOLO model family, which also covers segmentation, classification, depth, pose and tracking. For developers building computer vision apps.

Made by Ultralytics on Hugging Face. These notes are Mindket's, written from its model page and model card; the model's own files and license are on Hugging Face.

About

What it is: YOLO26 is an object detection model from Ultralytics, listed on Hugging Face for the ultralytics library. The company builds YOLO models for computer vision work, and the page says they cover detection, several kinds of segmentation, classification, depth, pose and tracking across video frames.

Who it suits: It suits developers and researchers who build vision features and want a fast, widely used detector with documentation in several languages.

Industry | Who uses it | How they use it | Example
Software development | App developers | Add object detection to an app that reads photos or video | Marking items found in an uploaded picture
Computer vision research | Researchers | Use it as a baseline and compare other detectors against it | Testing a new dataset on the same task
Video analysis | Engineers | Follow detected objects from one frame to the next | Tracking a moving object in a clip
Robotics and devices | Hardware teams | Run a detector on camera input | Spotting objects in a live camera feed

You can skip it if you need a chat model or text generation, since this one works on images and video.

Where to use it:

• Python projects that use the ultralytics library
• Windows, macOS or Linux with Python, to download the files
• Vision pipelines that take photos or video frames as input

Use case | Who | What they get back
Object detection | Developers | Boxes and labels for objects in an image
Segmentation | Researchers | Outlines of objects at the pixel level
Pose estimation | Engineers | Body or object keypoints for each detection
Tracking | Video teams | Detected objects followed across frames

How to set it up: You need Python and the Hugging Face download tool.

• Python on Windows, macOS or Linux
• The huggingface_hub package
• The ultralytics library for running the model

1. Install huggingface_hub.
2. Download the model files to your computer.
3. Follow the model card and the Ultralytics documentation to load them in the ultralytics library.

pip install -U huggingface_hub
hf download Ultralytics/YOLO26

How to use it: The record gives the download step, and the model card points to the Ultralytics documentation for running the model. After the download you load the files with the ultralytics library and pick the task you need.

hf download Ultralytics/YOLO26

Example: You download the files, load the model in a Python project and pass it a photo. For a detection task you would expect boxes and class labels for the objects it finds, as the card describes for its detection work.

For newcomers and experts: If you are new to computer vision, an object detector looks at a picture and draws boxes around the things it recognizes. A first small try is to download the files and run one photo through the library as the documentation shows.

For people who know the field, the family spans detection, instance and semantic segmentation, classification, depth, pose and oriented boxes, plus tracking over video. The card gives no speed or accuracy figures here, so check the paper it links before you compare it with other detectors.

Photo or video frame goes in
The YOLO26 model finds objects
Task output is boxes, masks or poses
Tracking links objects across frames

Good to know: The card text visible here is mostly links and badges, so the practical run steps live in the Ultralytics documentation.

• License: AGPL-3.0
• Cost to run: none named, though commercial use goes through a paid Enterprise License request
• Data it reaches: the images and video you give it, plus the downloaded model files
• Not for: commercial use under the open license alone, since the card tells commercial users to request an Enterprise License

On Hugging Face, 13 November 2026

• Task: object detection
• Library: ultralytics
• Downloads in the last 30 days: 26,767
• Likes: 264
• License: agpl-3.0
• Languages: en, zh, ja, ru, de, fr, es, pt
• Quantized versions: 60
• Fine-tunes: 140
• Tags: ultralytics, yolo, object-detection, instance-segmentation, image-classification, pose-estimation, tracking, semantic-segmentation, yolo26, model-index

Links

• Model: https://huggingface.co/Ultralytics/YOLO26
• Files: https://huggingface.co/Ultralytics/YOLO26/tree/main

Get the model

Download its files with Hugging Face's command line:

pip install -U huggingface_hub
hf download Ultralytics/YOLO26

Then follow the usage in its model card. Nothing in this kit runs by itself.