NVIDIA Open Models Power European Public Sector Projects
European governments and technology partners are deploying NVIDIA open models to build public-sector AI systems that improve services while keeping sensitive data, infrastructure and governance under local control. The effort centers on open, customizable models, locally relevant datasets and NVIDIA platforms that allow agencies to evaluate, adapt, secure and operate AI tools within their own environments. The European Commission is expanding deepset’s Haystack Enterprise Platform as the foundation of AI@EC, a shared environment for developing and running Commission AI applications on managed infrastructure. The update links agent development, execution, optimization and governance, adding capabilities to compare operating costs, evaluate answer quality and trace problems. NVIDIA Triton Inference Server supports shared search and ranking services, while reusable components, role-based access, audit trails and guardrails are intended to support document summarization, knowledge management and multilingual policy analysis. In Poland, Deloitte is supporting agentic AI projects across healthcare and employment services. The country’s eHealth Centre is building an intelligent imaging platform with NVIDIA DGX BasePOD and the MONAI framework to help physicians analyze X-rays, CT scans and MRI images, identify abnormalities and prioritize urgent cases. Separately, the Ministry of Family, Labour and Social Policy is developing an agentic factory on NVIDIA accelerated computing infrastructure to support more than 400 labor offices using PLLuM, GPT OSS and Inkling for document search, summarization and digital assistants. NVIDIA is also working with SpeakLeash in Poland and the University of Ljubljana in Slovenia on localized synthetic personas datasets. In the UK, Locai Labs released Nemotron-Personas-UK, a dataset of 1 million synthetic personas reflecting demographic and geographic diversity. The dataset is intended for bias testing, healthcare evaluation and public-service model validation across age groups, education levels, rural and urban populations. In Germany, Materna is introducing Sinfonie, an on-premises platform combining generative AI, voice interaction and 3D avatars, with NVIDIA Audio2Face available for animation. For classified environments, secunet, NVIDIA, Red Hat and deepset are developing a sovereign cloud based on secunet’s SINA Cloud foundation, combining NVIDIA accelerated computing, Nemotron open models, Red Hat OpenShift AI and Haystack orchestration in German data centers. The Soofi initiative is also producing Nemotron-Personas-Germany, a 1 million-record synthetic dataset grounded in census data for region and population testing. France and Greece are advancing local deployment. France’s CANUT selected LINAGORA and Docaposte to provide a sovereign collaborative workspace built on Twake.ai for more than 2,500 eligible organizations, including municipalities, schools and healthcare institutions. NVIDIA technology supports speech transcription, while infrastructure is hosted in France, with Région Occitanie as a preferred scaling partner. Greek AI and energy company KIEFER is developing an agentic factory on NVIDIA DGX SuperPOD and DGX GB200 using Sophea open-weight models and Greek-language tools for knowledge retrieval, speech and digital assistants. NVIDIA’s Open Models Codefest is supporting startups, software providers and labs across Europe, with top teams Pleias, Zehnlab.ai, Disseqt, Algoritmica GmbH and ABA Technology, while Deutsche Telekom and T-Systems have provided compute infrastructure.
