Meet YOLO26: next-gen vision AI.
Ultralytics
Roadmap

Where Ultralytics is headed

From the models that brought real-time vision AI to millions, to the releases on the horizon. Here's what we've shipped and what's coming next.

ShippedOpen source
2020

The release that brought real-time object detection to millions — PyTorch-native, fast, and remarkably easy to train.

ShippedMulti-task
2023

One framework for detection, segmentation, classification, pose estimation, and oriented bounding boxes.

ShippedEfficiency
2024

A refined architecture delivering higher accuracy with fewer parameters, unveiled at YOLO Vision 2024.

ShippedLatest stable
2025

Our current recommended model — faster, more accurate, and production-ready across every vision task.

ShippedPlatform
March 2026

The end-to-end platform to annotate data, train YOLO models, and deploy across 43 global regions — all in one place.

ShippedSegmentation
May 2026

Our first semantic segmentation models — dense per-pixel class labels for full-scene understanding.

Today
Tracking
June 2026
Improved trackers

Faster, more accurate multi-object tracking — more stable identities through occlusions and crowded scenes for real-world video.

  • Re-ID — re-identification keeps object identities consistent across cameras and after occlusions
Training
June 2026
Knowledge distillation

Built-in distillation to compress large teacher models into smaller, faster students that keep their accuracy — ideal for edge and real-time deployment.

Anomaly detection
July 2026
YOLO-Anomaly

Purpose-built anomaly detection for manufacturing quality assurance — catching defects and out-of-spec parts right on the production line.

Next-gen flagship
September 2026
YOLO27

Shenzhen, China · YOLO Vision 2026

The next flagship generation of YOLO, unveiled live at YOLO Vision 2026, expanding the family into 3D perception:

  • YOLO-Depth — monocular depth estimation from a single camera
  • YOLO-StereoDepth — binocular disparity depth for robotics, a camera-native alternative to lidar
Platform
Late 2026
Platform AutoTrain

Iterative, LLM-driven training analysis that automatically diagnoses each run and refines the configuration over successive rounds to push accuracy higher.

On the horizon
2027
New modalities

New capabilities join the YOLO family throughout 2027:

  • YOLO-OCR — fast, accurate text recognition
  • YOLO-Face — facial recognition and analysis
  • YOLO-VLM — a lightweight YOLO front-end feeding a deeper LLM layer for efficient vision-language pipelines

Build on the latest YOLO

Start training and deploying with YOLO26 today — and be ready for everything that's next.