Aleph Alpha, a Heidelberg-based company, has launched Kolibri, an artificial intelligence model with 78.1 billion parameters and open weights. This Mixture-of-Experts model features a context window of up to 1 million tokens and is tuned for English and German. Trained on 20 to 24 trillion tokens with particular emphasis on German content, it was developed using 768 NVIDIA B200 GPUs located in Germany and Finland. The model is available under the Apache 2.0 license on Hugging Face, requiring approximately 78 GB of storage in FP8 format. Aleph Alpha targets regulated sectors including public administration, industry, aerospace, and defense.
Source: Read the original article
Disclaimer: this content is for information purposes only and is not financial advice. Cryptocurrencies are highly volatile: you may lose all of your capital. Always do your own research. Legal notice
