Real-time YOLO object tracking
Follow every object through video with persistent IDs. Ultralytics YOLO26 tracking combines real-time detection with proven trackers, powering counting, analytics, and monitoring on live streams.
Tracking runs on YOLO26
Tracking adds persistent IDs on top of any YOLO26 model — one method call turns a detector into a multi-object tracker with BoT-SORT or ByteTrack. Detector speed sets tracking throughput, so the benchmarks below are the numbers that matter for live streams.
pip install ultralyticsfrom ultralytics import YOLO
# Load a YOLO26 model
model = YOLO("yolo26n.pt")
# Run inference
results = model.track("video.mp4")| Model | Size (px) | mAP 50-95 | mAP 50-95 (e2e) | CPU ONNX (ms) | T4 TensorRT (ms) | Params (M) | FLOPs (B) |
|---|---|---|---|---|---|---|---|
| YOLO26n | 640 | 40.9 | 40.1 | 38.9 ± 0.7 | 1.7 ± 0.0 | 2.4 | 5.4 |
| YOLO26s | 640 | 48.6 | 47.8 | 87.2 ± 0.9 | 2.5 ± 0.0 | 9.5 | 20.7 |
| YOLO26m | 640 | 53.1 | 52.5 | 220.0 ± 1.4 | 4.7 ± 0.1 | 20.4 | 68.2 |
| YOLO26l | 640 | 55.0 | 54.4 | 286.2 ± 2.0 | 6.2 ± 0.2 | 24.8 | 86.4 |
| YOLO26x | 640 | 57.5 | 56.9 | 525.8 ± 4.0 | 11.8 ± 0.2 | 55.7 | 193.9 |
YOLO26 detection benchmarks from the Ultralytics docs; trackers run on these models.
Trusted by the world's leading organizations
Our models' impact
Streamline processes across industries with our cutting-edge vision AI models. Speed, accuracy and ease-of-use powered by Ultralytics.
The evolution of Ultralytics YOLO models
See how Ultralytics YOLO evolved from the practical YOLOv5 workflow to edge-ready YOLO26 inference.
Label up to 10x faster with smart annotation
Ultralytics Platform gives you the image annotation tool to build high-quality datasets faster. From smart annotation to precise manual editing, these features are designed to reduce image labeling time without sacrificing quality.
- SAM-powered smart annotation: Masks and bounding boxes in one click.
- Full AI task coverage: Detection, instance segmentation, semantic segmentation, classification, pose, OBB.
- Universal format support: Your choice of YOLO, COCO, and more.
- Team review and versioning: Clear collaboration at every step.
Train on the Best GPUs for Less
26 NVIDIA GPUs starting at $0.24/hr — from Ampere to Blackwell. No markup, no minimums, no commitment.
Explore industry solutions
See how teams apply Ultralytics computer vision across production environments.
Agriculture
Automotive
Healthcare
Logistics
Manufacturing
Retail
Robotics
Agriculture
Automotive
Healthcare
Logistics
Manufacturing
Retail
Robotics
Agriculture
Automotive
Healthcare
Logistics
Manufacturing
Retail
Robotics
Frequently asked questions
Multi-object tracking detects objects in every video frame and links them across frames, assigning each object a persistent ID. That identity over time is what enables counting unique objects, measuring dwell time, analyzing trajectories, and triggering events when a specific object crosses a line or enters a zone.
Ultralytics ships BoT-SORT and ByteTrack out of the box, selectable with a single argument in track mode. Both pair the detector's outputs with motion-based association; BoT-SORT is the default, while ByteTrack is a strong lightweight alternative. Tracker behavior is configurable through YAML files.
Tracking runs on top of any Ultralytics detection, segmentation, pose, or OBB model. Ultralytics YOLO26 is the recommended base: its real-time speed keeps tracking smooth on live streams, and the nano size makes edge tracking practical.
Yes. Train a detector on your own classes with train mode or on Ultralytics Platform, then call track mode with your trained weights. The tracker follows whatever your model detects, so custom tracking needs no extra tracker training.
Typical applications include people and vehicle counting, retail and venue analytics, sports player tracking, traffic monitoring, and safety systems that follow objects through restricted zones. Because Ultralytics tracking runs in real time, these systems work on live camera feeds rather than only recorded video.
Build object tracking with Ultralytics
Train YOLO26 models and track objects with persistent IDs in real time using Ultralytics Platform.