A technical article on Kolibri, Aleph Alpha's sovereign language model, examining architectural trade-offs in autoregressive models including parameter allocation, FLOPs scaling, and sequence-mixer state. The piece provides an interactive tool for configuring and comparing model architectures against recent open-weight releases.
Aleph Alpha released Kolibri, an open-weight bilingual English-German model with 78.1B parameters and 1M token context window, designed for sovereign deployment in government and regulated industries. The Mixture-of-Experts model emphasizes grounding through abstention training and achieves strong performance on math, coding, and long-context tasks while maintaining lower inference costs than larger competitors.