German AI consortium releases Soofi S, an open 30B model that tops benchmarks in both English and German
Researchers have released Soofi S, an open-source, hybrid foundation model designed for proficiency in both German and English. By combining Mamba architecture with a Mixture-of-Experts approach, the 30-billion parameter model activates only 3 billion parameters per token to achieve consistent inference speeds. This release provides a new resource for developers seeking a sovereign alternative to existing large language models.
Covered by 2 sources
- TThe Decoder↗Jonathan KemperJul 13
- AarXiv CS.AI↗The Soofi-Team, :, Benedikt Droste, David Fitzek, Ruben H\"arle, Lukas Helff, Maximilian Idahl, Alex Jude, Abbas Goher Khan, Maurice Kraus, Timm Ruland, Richard Rutmann, Sebastian Sztwiertnia, Markus Frey, Daniil Gurgurov, Jan Pfister, Tom R\"ohr, Sebastian von Rohrscheidt, J\"org Bienert, Nicolas Flores-Herr, Simon Gottschalk, Andreas Hotho, Kristian Kersting, Joachim K\"ohler, Alexander L\"oser, Wolfgang Nejdl, Simon Ostermann, Jan Plogsties, Patrick Putzky, Mehdi Ali, Michael Fromm, Max L\"ubberingJul 13