A group of Swiss institutions has released a new open AI model, designed to serve as a foundation for future research and applications. Built by EPFL, ETH Zurich, and the Swiss National Supercomputing Centre (CSCS), the model is called Apertus – Latin for “open.” The name reflects its core principle: every part of its design and training process is accessible to the public.
Developers and organisations can use Apertus to create chatbots, translation tools, or education-focused applications. It can be downloaded directly from Hugging Face or accessed through Swisscom, a strategic partner of the initiative. Two versions are available – an 8-billion-parameter model and a larger 70-billion-parameter version. Both are released under a permissive open-source licence, allowing use in research, education, and commercial projects.
Built for openness
Unlike other AI systems that reveal only select details, Apertus is a fully open AI model, with its architecture, training data, and documentation available for inspection.
“With this release, we aim to provide a blueprint for how a trustworthy, sovereign, and inclusive AI model can be developed,” said Martin Jaggi, Professor of Machine Learning at EPFL and member of the Steering Committee of the Swiss AI Initiative. He said Apertus will be updated regularly by a team of engineers and researchers from CSCS, ETH Zurich, and EPFL.
Thomas Schulthess, Director of CSCS and Professor at ETH Zurich, described Apertus as “a driver of innovation and a means of strengthening AI expertise in research, society and industry.” He said the project is not a typical transfer of technology from research to product, but an effort to build infrastructure for long-term use.
Multilingual reach
The training process involved 15 trillion tokens in more than 1,000 languages, with about 40% of the data in non-English languages. Apertus includes languages often left out of LLMs, like Swiss German and Romansh.
“Apertus is built for the public good. It stands among the few fully open LLMs at this scale and is the first of its kind to embody multilingualism, transparency, and compliance as foundational design principles,” said Imanol Schlag, technical lead of the project and Research Scientist at ETH Zurich.
Swisscom is already deploying Apertus on its sovereign AI platform. “This underscores our commitment to shaping a secure and responsible AI ecosystem that serves the public interest and strengthens Switzerland’s digital sovereignty,” said Daniel Dobos, Research Director at Swisscom.
Testing the open AI model: Access and real-world use
While downloading Apertus is simple for experienced users, practical use requires servers, cloud resources, or dedicated interfaces. Developers will be able to test Apertus during the Swiss {ai} Weeks which continue until October 5, 2025. Hackathon participants will gain access through a Swisscom-hosted interface. Swisscom business customers can also begin using the model today through the company’s AI platform. For international users, Apertus will be available through the Public AI Inference Utility.
“Currently, Apertus is the leading public AI model: a model built by public institutions, for the public interest. It is our best proof yet that AI can be a form of public infrastructure like highways, water, or electricity,” said Joshua Tan, Lead Maintainer of the Public AI Inference Utility.
Transparency and compliance
Under the open-source licence training data, model weights, and intermediate checkpoints are available. The model’s training process followed Swiss data protection rules, Swiss copyright law, and the transparency requirements of the EU AI Act.
The dataset was limited to publicly-available information, filtered to remove personal data and to honour website opt-out requests. Ethical guidelines were also applied to exclude unwanted material before training began.
The future of Switzerland’s open AI model
“Apertus demonstrates that generative AI can be both powerful and open,” said Antoine Bosselut, Professor at EPFL and Co-Lead of the Swiss AI Initiative. “The release of Apertus is not a final step, rather it’s the beginning of a journey, a long-term commitment to open, trustworthy, and sovereign AI foundations, for the public good worldwide.”
Future updates aim to expand the model family, improve efficiency, and develop domain-specific tools for areas like law, health, climate, and education – while continuing to uphold strict standards of transparency.
Source: https://www.artificialintelligence-news.com/