Partners
Built with a machining laboratory, on open standards.
We list relationships that exist and standards we actually build on. Product names on this site are trademarks of their owners and do not imply endorsement.
Research & validation
KSF, Tuttlingen
The machining competence centre whose laboratory hosts Toleron's development fleet and whose GrindingExpert service is integrated in Grind Master.
Toleron is developed on the laboratory’s machines — grinding, turning and milling, on Siemens Sinumerik and Haas controls and on a sensor-only grinder. The proof points on this website come from work there.
The laboratory’s grinding expertise is inside the product: Grind Master integrates the KSF GrindingExpert service, which returns complete dressing and grinding strategies with warning levels for a given material, wheel and target quality.
Open standards & technologies
What the platform is built on.
No proprietary interfaces, no lock-in at the controller. This is what Toleron is built on.
Open controller standards
Machine data read straight from the controller, read-only — including alarms
Siemens and Haas, validated
Sinumerik 840D sl and 828D, Sinumerik Industrial Edge, Haas Next Generation Control
Open model format
AI models run at the machine, on the AI box or on the central server
Open-source foundations
Proven open-source components for data storage and local AI
Working with Toleron
Three ways in.
Manufacturers
Connect one machine, validate on real parts, then roll out to more machines without re-engineering each one. Runs in your plant — your data stays on your network.
Machine builders & integrators
Ship monitoring with the machine or retrofit it in the field. Open standards, documented data access and evaluations your own engineers can build themselves.
Research groups
Detailed, time-aligned machine and sensor data with a permanent archive and honest exports — the data foundation for process research.
Next step
Bring intelligence to your machines.
Connect existing CNC equipment without replacing the machine or the controller. We start with one machine, real data and an honest acceptance run.