ETH Zurich, EPFL and CSCS released Apertus 1.5 on 24 July 2026. The model is trained on the Alps supercomputer in Lugano and published under Apache 2.0, with training data, code and development process open to inspection.
It now reads images and audio too, reasons better, follows instructions more closely and handles tool use more reliably. This opens the door to a broader range of applications, and to a question: who is going to build them?
Together with over 85 organisations, aiLights is part of a joint community effort to help accelerate the adoption of Apertus:
It’s called «Hack Apertus».
We mobilise people to build open-source blueprints for embedded solutions across sectors, industries and regions, on this sovereign foundation.
Over two weeks, teams get hands-on with Apertus, Switzerland’s open large language model. In the Use Case track, they build prototypes that put Apertus to work across industries. In the Academia track, university cohorts examine the model closely, testing its capabilities, its grasp of Swiss dialects, or bringing their own data and use case to prototype a domain application. Findings from Stage 1 carry forward into Stage 2, the 6+ On-site Hackathons across Switzerland.
«Hack Apertus» runs from Oct ’26 until May ’27 in three connected stages
Online Hackathon: 1–16 Oct ’26
On-site Hackathons: Oct ’26 to May ’27, in 6+ Swiss cities
Grand Finals: 14 May ’27 in St.Gallen/HSG
As part of the Swiss AI Action Plan, «Hack Apertus» feeds directly into Switzerland’s positioning ahead of the Geneva AI Summit 2027, contributing open, trustworthy and sovereign AI assets that demonstrate the country’s leadership in responsible AI innovation.
If you are interested in building the bridge from world class research to industry & public sector use, you are warmly invited!
Join us and sign up to be part of the effort.
You need to load content from reCAPTCHA to submit the form. Please note that doing so will share data with third-party providers.
More InformationYou are currently viewing a placeholder content from Turnstile. To access the actual content, click the button below. Please note that doing so will share data with third-party providers.
More InformationYou are currently viewing a placeholder content from Vimeo. To access the actual content, click the button below. Please note that doing so will share data with third-party providers.
More InformationYou are currently viewing a placeholder content from YouTube. To access the actual content, click the button below. Please note that doing so will share data with third-party providers.
More InformationYou need to load content from reCAPTCHA to submit the form. Please note that doing so will share data with third-party providers.
More Information