German AI startup Aleph Alpha has newly announced its open-weight model, "Kolibri." This model is part of the company's strategy to establish "AI sovereignty" in Europe, featuring high transparency and reliability, and is designed specifically for use by enterprises and public institutions that demand rigorous security.
Kolibri is an open-weight model that applies the development expertise Aleph Alpha has cultivated through its commercial LLMs. Users are able to control and operate the model within their own data environments, achieving efficient and high-precision inference tailored to specific operational needs.
In the European market, strict data privacy regulations such as GDPR are paramount. Kolibri aims to technically underpin "AI sovereignty" by minimizing reliance on external cloud infrastructure and supporting AI adoption that complies with European regulations. By making the model's architecture transparent, it provides a highly reliable solution to the biggest concern in enterprise AI adoption: the black-box nature of AI.
Aleph Alpha will continue to build environments where industries and the public sector within Europe can safely and advancedly utilize AI. Through this open approach, the company plans to strengthen the independence and competitiveness of the European AI ecosystem.