Model weights · YOLOv5 (v5u)

Download YOLOv5 (v5u) weights

The anchor-free v5u retrain of the classic PyTorch YOLO, keeping its easy training and export while adopting the newer YOLOv8 detection head for better accuracy. Pick a size below and download the official .pt checkpoint in one click, with published COCO accuracy, size and license all in view.

AGPL-3.0Ultralytics · 2020 · Ultralytics (PyTorch)

YOLOv5 (v5u) checkpoints, pick a size and download

The anchor-free v5u retrain of the classic PyTorch YOLO, keeping its easy training and export while adopting the newer YOLOv8 detection head for better accuracy.

  • YOLOv5nu
    yolov5nu.pt
    COCO mAP
    34.3
    Params
    2.6M
  • YOLOv5su
    yolov5su.pt
    COCO mAP
    43.0
    Params
    9.1M
  • YOLOv5mu
    yolov5mu.pt
    COCO mAP
    49.0
    Params
    25.1M
  • YOLOv5lu
    yolov5lu.pt
    COCO mAP
    52.2
    Params
    53.2M
  • YOLOv5xu
    yolov5xu.pt
    COCO mAP
    53.2
    Params
    97.2M

Scores are COCO mAP at 640px, published by the authors. Weights host: github.com. Clicking Download verifies the file and starts it straight from the official CDN.

How to load YOLOv5 (v5u) weights

Install the Ultralytics package, then point the loader at the checkpoint. It downloads automatically on first use, or you can pass the local path to the file you downloaded above.

pip install ultralytics
from ultralytics import YOLO

# Downloads on first use, or pass a local path to your .pt file
model = YOLO("yolov5nu.pt")
results = model("image.jpg")

Want to see the architecture? Export to ONNX and open it in the ONNX visualizer for a labelled diagram with shapes and parameter counts.

Open visualizer →

Choosing a YOLOv5 (v5u) checkpoint

Read off the table above: what each step up the size ladder costs and returns for this family specifically, where it stops being worth it, and what the license actually permits.

What the size ladder buys

YOLOv5 (v5u) spans YOLOv5nu at 2.6M parameters to YOLOv5xu at 97.2M, 37.4× the model, for 18.9 more COCO mAP (34.3 to 53.2). That works out to roughly 0.20 points per additional million parameters across the whole range, and the return is front-loaded: the early steps are much cheaper than the last one. Published figures from Ultralytics at 640px, not re-measured here.

Where the ladder stops paying

The best value step is YOLOv5nu → YOLOv5su, worth 1.34 COCO mAP points per million parameters. The worst is YOLOv5lu → YOLOv5xu at 0.02, or 59× less efficient, for 44.0M extra parameters and only 1.0 more points. If you are latency- or memory-bound, that is the step to skip; if you are accuracy-bound and the compute is free, it is the only place left to get it.

Where it sits in the catalog

Against the 8 families here that publish COCO mAP, YOLOv5 (v5u)'s best checkpoint ranks 8th at 53.2, against 57.5 for YOLO26. Read that as a tier indicator rather than a head-to-head: the families use different size ladders, input sizes and training recipes, and each number is the authors' own. The closest alternatives for object detection are YOLO26, YOLO11 and YOLO12.

Licensing, in practice

The weights are AGPL-3.0. AGPL-3.0 is free for open-source, research and internal use; a closed-source commercial product needs an Ultralytics Enterprise license.

About YOLOv5 (v5u)

YOLOv5 made YOLO practical for everyone: pure PyTorch, easy training and export, and a huge deployment footprint that persists today. The download files below are the anchor-free YOLOv5u retrain, which adopts the YOLOv8 detection head for higher accuracy while keeping the familiar n/s/m/l/x ladder. The original anchor-based YOLOv5 releases remain available on the ultralytics/yolov5 repo.

Author
Ultralytics
Released
2020
Tasks
Detect, Segment, Classify
Framework
Ultralytics (PyTorch)
Input size
640px
License
AGPL-3.0
Resources
DocsGitHubLicense

More model weights to download

Each page lists every checkpoint with accuracy, parameters, license and a one-click download.

See all 19 model families →

YOLO tracker FPS benchmarks →Get in touch →
Blog

From the blog

Tutorials, code, and notes on computer vision, deep learning, and applied AI.

View all posts →

Frequently asked questions

How do I download YOLOv5 (v5u) weights?
Click the Download button next to any variant in the table above; we verify the file and start it straight from the official CDN. You can also let the loader fetch it automatically on first use with `model = YOLO("yolov5nu.pt")`. All 5 YOLOv5 (v5u) checkpoints are hosted officially.
Is YOLOv5 (v5u) free for commercial use?
YOLOv5 (v5u) is released under AGPL-3.0. That is free for open-source and research; closed-source commercial deployments need an Ultralytics Enterprise license. Always confirm against the linked license text.
Which YOLOv5 (v5u) model size should I use?
Start with YOLOv5nu, the smallest and fastest, ideal for prototyping, edge and CPU. Move up the ladder only when you need more accuracy and have the compute for it. The largest variant reaches 53.2 COCO mAP.