Aleph Alpha launches Kolibri-1, an open model with 78 billion parameters and up to 1 million tokens of context
In 30 seconds
German AI company Aleph Alpha has released Kolibri, a language model with open weights that can be downloaded from Hugging Face and used under the Apache 2.0 license. It has 78 billion parameters, of which only 3 billion are active at each step, and supports up to 1 million tokens of context. It only works in German and English, and it needs powerful hardware, such as two 80 GB A100 GPUs or one H200.
German AI company Aleph Alpha released Kolibri, its new language model, on October 3, German Unity Day. The weights are open: the full model can be downloaded from Hugging Face and used under the Apache 2.0 license.
It is a mixture-of-experts model (it splits the work across specialized parts): 78 billion parameters in total, but only 3 billion active at each step. According to the company, this keeps the computing it needs low, although the full model has to be held in memory.
It supports up to 1 million tokens of context (the amount of text it can take into account at once), although the company recommends staying at about 262,000 or below for efficiency and for complex tasks.
It can reason step by step and call external tools. According to Aleph Alpha, in math, coding and long texts it matches models with up to four times as many active parameters.
It is built for public administration, industry and aerospace, and to run on in-house servers without sending internal data to third parties.
One caveat: it only works in German and English, and it needs powerful hardware, such as two 80 GB A100 GPUs or one H200.
We see it as another step for open European AI, although today, for most small businesses, it is more of a reference point than an everyday tool.
Why it matters
It can run on in-house servers without sending internal data to third parties, but only in German and English.
Official source: Aleph Alpha
