# Broken Yolo Demo project in the PyTorch Docs

**URL:** <https://community.ultralytics.com/t/broken-yolo-demo-project-in-the-pytorch-docs/2292>\
**Category:** Ultraltyics Platform\
**Created:** [November 2, 2026, 12:31am UTC](https://community.ultralytics.com/t/broken-yolo-demo-project-in-the-pytorch-docs/2292 "2026-11-02T00:31:07Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![eoinedge](https://sea1.discourse-cdn.com/flex001/user_avatar/community.ultralytics.com/eoinedge/32/1802_2.png) [@eoinedge](https://community.ultralytics.com/u/eoinedge)\
**Post date:** [November 2, 2026, 12:31am UTC](https://community.ultralytics.com/t/broken-yolo-demo-project-in-the-pytorch-docs/2292/1 "2026-11-02T00:31:07Z")

</div>

I’m working on a demo for an ExecuTorch guide and wanted to use this model. Its featured in the PyTorch docs, is there any way to bring the demo back?

> **[YOLOv5 - PyTorch](https://pytorch.org/hub/ultralytics_yolov5/)**
>
> Ultralytics YOLOv5 🚀 for object detection, instance segmentation and image classification.

 ![image](https://us1.discourse-cdn.com/flex001/uploads/ultralytics1/original/2X/4/44c37bccd58b12f7bfd00dd9dff66a46a66c6c01.png)

(Link to the demo space is here: [https://platform.ultralytics.com/projects/DnIqQCdlI37oliWBAoQJ](https://platform.ultralytics.com/projects/DnIqQCdlI37oliWBAoQJ))

---

<div class="post-metadata">

**Author:** ![Toxite](https://sea1.discourse-cdn.com/flex001/user_avatar/community.ultralytics.com/toxite/32/123_2.png) [@Toxite](https://community.ultralytics.com/u/Toxite)\
**Post date:** [November 2, 2026, 8:04am UTC](https://community.ultralytics.com/t/broken-yolo-demo-project-in-the-pytorch-docs/2292/2 "2026-11-02T08:04:17Z")

</div>

Hello @eoinedge

YOLOv5 linked in that repo is our legacy model. You can find a similar, but not identical version of it, here:

> **[YOLOv5 Models by Ultralytics](https://platform.ultralytics.com/ultralytics/yolov5)**
>
> Released June 2020. The foundational PyTorch implementation, battle-tested across millions of deployments. Ideal for teams with existing YOLOv5 workflows.

The output format of this isn’t identical as legacy YOLOv5, however ExecuTorch export is natively supported through that website ([Yolov5nu Model by Ultralytics](https://platform.ultralytics.com/ultralytics/yolov5/yolov5nu?tab=export)) or through Ultralytics library.

> **[Usage - YOLO26 ExecuTorch for Mobile & Edge | Ultralytics Docs](https://docs.ultralytics.com/integrations/executorch#usage)**
>
> Deploying computer vision models on edge devices like smartphones, tablets, and embedded systems requires an optimized runtime that balances performance with resource constraints. ExecuTorch, PyTorch's solution for edge computing, enables efficient...