# Extending YOLO26 for custom multi-task architecture
**URL:** https://community.ultralytics.com/t/extending-yolo26-for-custom-multi-task-architecture/1811
**Category:** Discussion
**Tags:** question
**Created:** [February 12, 2026, 2:13pm UTC](https://community.ultralytics.com/t/extending-yolo26-for-custom-multi-task-architecture/1811 "2026-02-12T14:13:51Z")
**Posts on this page:** 2
**Page:** 1
### Author:  [@carlos](https://community.ultralytics.com/u/carlos)
#### Post date: [February 12, 2026, 2:13pm UTC](https://community.ultralytics.com/t/extending-yolo26-for-custom-multi-task-architecture/1811/1 "2026-02-12T14:13:51Z")
Hi,
I would like to confirm my understanding of YOLO26’s architectural flexibility.
My goal is to build a single model that:
- Detects one class (“car”) using bounding boxes
- Performs segmentation on different classes (“door”, “window”, “damage”)
From my understanding, the standard YOLO26 segmentation configuration ties segmentation classes to detection classes, which would not support this setup directly.
Therefore, I believe the appropriate approach would be to:
- Use the YOLO26 backbone + neck as a shared encoder
- Attach independent task heads:
This would result in a true multi-task architecture with separate label spaces per head.
Could you please confirm whether this interpretation is correct, or if there is a simpler way to achieve this using YOLO26 without modifying the architecture?
Thank you so much
---
### Author:  [@Toxite](https://community.ultralytics.com/u/Toxite)
#### Post date: [February 12, 2026, 3:35pm UTC](https://community.ultralytics.com/t/extending-yolo26-for-custom-multi-task-architecture/1811/2 "2026-02-12T15:35:47Z")
You could with significant changes in code since it would break a lot of Ultralytics assumptions.
Easier option is to have a mask for the whole car and train with `overlap_masks=False` so you can get the masks for other classes too.