Object detection on photos
Boxes drawn on equipment, anchors or damage in a field photo.
AI Development
YOLO development services and the licence question
We have no published YOLO build. We have shipped photo-evidence inspection apps, and we show where object detection would sit and what the Ultralytics licence means for a commercial product.
0Published builds on YOLO
YOLO26Latest model per the Ultralytics docs
YOLO11Previous generation
2Licence options: AGPL-3.0 or Enterprise
What we build
What we would take on with YOLO
Segmentation and pose
The other tasks the Ultralytics package supports.
A licence decision first
AGPL-3.0 or the Enterprise License, settled before the build starts.
Rules around the result
A detection feeds validation rules and a certificate before anyone calls it a verdict.
Architecture
Where YOLO would sit
Detection between the photo and the record
Field app
Photo and signatureCaptured on site
Detection
YOLO modelWhere it would sit
FastAPIPython 3.13 service
Record
Validation rulesThresholds and exceptions
PDF certificateGPS, timestamps, readings
Illustration. A dashed outline marks where the technology would sit. We have no published build on it.
Licence
AGPL-3.0 or the Enterprise License
The two Ultralytics licence routes
Open source
AGPL-3.0
- An OSI-approved licence the repository describes as suited to students, researchers and enthusiasts
- Carries open-source requirements
Commercial
Ultralytics Enterprise License
- For commercial development and production use
- Avoids the open-source requirements of AGPL-3.0
Source: Ultralytics repository on GitHub and Ultralytics documentation, checked 4 November 2026. Ask Ultralytics Licensing, or your lawyer, which route fits your product.
Builds
The closest builds we can show you
Statutory inspection, United Kingdom
Work-at-height anchor and torque platform
Photo evidence from the field with validation rules and certificates on completion.
70%Faster reporting cycle
Logistics, fleet operations
TruckFleet Labs
A photo and a signature before a drop closes, on a native iOS app that works with no signal.
6 to 1Disputes a month, from the photo requirement alone
Fit check
When YOLO is the right call
Pick YOLO when
- You need to find objects in photos or video
- Speed matters more than a perfect box
- Your team can settle the licence route first
Look elsewhere when
- You need a published YOLO build to inspect, which we do not have
- You cannot accept AGPL-3.0 and will not buy the Enterprise License
- A human can check every photo
Explore
Same job, other tools
Pairs with
Where we used it
Stack
What YOLO usually sits beside
OpenCV
PyTorchPython
Docker
TensorFlow
KubernetesFAQs
What buyers ask first
Which YOLO version is current?
The Ultralytics documentation names YOLO26 as the latest and YOLO11 as the previous generation. It says YOLO27 has no launch date yet.
Can we use YOLO in a commercial product?
The Ultralytics repository offers AGPL-3.0 or a commercial Enterprise License that avoids AGPL-3.0 open-source requirements. Settle the route with Ultralytics before you build.
Do you have a published YOLO build?
No. We say so on the first call and show the nearest work, which is photo-evidence capture on inspection apps.
Who owns the code and the cloud accounts?
You do, from the first commit. Repositories and cloud accounts stay in your name.
Tell us what your detector has to find
Thirty minutes on your images, your licence needs and your rules. No sales pitch.