Founded in 2025 in Cagliari by Andrea Pili, Michele Fadda and Francesca Audino, Xference.ai offers a distributed platform that lets enterprises run open-source LLMs such as Qwen, Llama and Mistral entirely inside their own servers with no internet connectivity or data off-loading. The system couples an optimized inference engine, a reasoning layer, no-code tools and plug-and-play API connectors so that non-technical users can query internal knowledge bases securely. Its architecture is designed to scale linearly from a single node to thousands, minimize latency, cut power consumption and remain compatible with commodity CPUs as well as AI accelerators. Xference positions itself as “privacy by design,” natively aligned with GDPR and the EU AI Act, and has already deployed its first infrastructure in a major Italian data center as part of a European network. The company will soon launch an invitation-only beta for 100 early adopters. Financially, it has secured €700 000 in pre-seed funding and is now seeking a larger seed round to accelerate product development, expand the team and capture enterprise demand for private AI inference across Europe.

Total raised

$1M

Funding rounds

1

Latest round

Seed

Latest activity

Feb 2026