Aleph Alpha's Kolibri: A 78B-Parameter Open Model That Hits the Pareto Frontier in English and German
Kolibri – Tech Report [pdf]
Aleph Alpha released Kolibri, an English–German Mixture-of-Experts transformer with 78.1B total parameters and 3.46B active per token, under Apache 2.0. Trained on 24T tokens—over 20% German—and post-trained with RL on 1.2M tasks, it targets sovereign and regulated domains like public administration, industry, and aerospace. Among open models, it sits on the Pareto frontier of quality and serving cost in both languages.
Kolibri builds on Kolibri Origin, an earlier iteration that validated our full pipeline at scale.
- senko
Since it's a PDF, here's an excerpt from the summary:
> This report introduces Kolibri, an English–German Mixture-of-Experts transformer with 78.1B total parameters and 3.46B
active parameters per token, released as open weights under the Apache 2.0 licence.
> We built Kolibri for sovereign and
specialised deployment, with a particular focus on German and on regulated domains such as public administration, industry,
and aerospace. To support these settings, we control the full model-development process, including data, architecture,
training infrastructure, post-training, and evaluation.