Cost calculator for on-premises AI: developing a PoC
With local AI, your data stays with you. A proof of concept puts your use case into production-like operation in your own environment.
But how much effort does that take? Describe your use case and get an initial effort estimate in person-days within two minutes. The calculator is free and requires no sign-up.
How the estimate works: Every proof of concept starts from a base, the inference stack including basic setup and a standard frontend. On top of that, you select the building blocks your use case requires, such as connecting your own data sources or single sign-on. The effort figures are ranges drawn from our project experience, and the estimate below updates with every selection.
e.g. file server / NAS / file system, wiki, SharePoint, websites. Includes chunking and embedding pipeline.
Connection to ERP, ticketing, e-mail or other leading systems.
Sign-in via your existing directory service, e.g. Entra ID or LDAP.
Permissions are taken over from the respective source (file system permissions, SharePoint permissions, etc.). Scales with the number of data sources.
Test data set, quality metrics and structured feedback from pilot users. What separates a reliable PoC from a mere installation.
A custom interface instead of the standard frontend. For a PoC, the standard frontend is usually sufficient.
GDPR documentation and EU AI Act classification for the pilot phase.
Contact Us
Case Studies and Use Cases
Contact Us
Are you ready to launch your own AI project? We’d be happy to provide a customized quote for your company’s AI solution or advise you on your options and our packages.