CloudFerro Clouds and Sherlock Platform

CloudFerro Sherlock is a secure Inference-as-a-Service platform for generative AI, developed within the Gateway IPCEI-CIS project. It lets organisations deploy large language models on sovereign European cloud infrastructure without building the AI stack themselves. Models are available through an OpenAI-compatible API, so integration is simple. They include Bielik v3.0, GPT-OSS 120B, PLLuM-8x7-chat, LLaMA 3.1 and 3.3, MiniMax M2.5 and Pixtral. Typical use cases are knowledge assistants, document analysis, advanced search, coding support and Polish-language applications. How it connects to Ambiti8n: In the Ambiti8n project, Sherlock provides a trusted inference path for models that are too large to run at the hospital edge and too sensitive for generic external AI APIs. CloudFerro's cloud also hosts the hospital-side environments used for integration and testing in the demonstrator. Sensitive data stays under control, while applications reach advanced AI through a secure, standard API. The Villanova model from a consortium partner has also joined the Sherlock model catalogue, so the cooperation outlasts the project. Sherlock shows how sovereign, federated and practical generative AI can work across the European cloud-edge ecosystem, in line with the values of IPCEI-CIS and 8ra: open collaboration, interoperability and digital sovereignty. https://ambiti8n.eu/pages/blog/no7-cl... www.cloudferro.com