Meet LeoLM: The First Open and Commercially Available German Foundation Language Model
Built on WEB and Trained on a Large-Scale German Text Corpus
Introducing LeoLM-7B, 13B, and 70B
We are proud to introduce LeoLM, the first open and commercially available German Foundation Language Model. Built on WEB and trained on a large-scale high-quality German text corpus, LeoLM sets a new standard for German language processing.
LeoLM is available in three sizes: 7B, 13B, and 70B. With continued pretraining on German texts, we are excited to offer LeoLM-70B in the near future.
LeoLM-7B and 13B are now available for use. These models are ideal for a wide range of natural language processing tasks, including:
- Text generation
- Machine translation
- Question answering
- Text summarization
- Chatbots
LeoLM is also available as a finetuned version for German instructions and conversations in the style of Open Assistant tokens.
German LLMs Based on LLaMA-2 Pre-Training
LeoLM is based on LLaMA-2, a large language model developed by Meta AI. LLaMA-2 has been shown to achieve state-of-the-art performance on a wide range of natural language processing tasks.
By building LeoLM on LLaMA-2, we have created a German language model that is powerful, versatile, and easy to use.
Get Started with LeoLM Today
To get started with LeoLM, please visit our website. We offer a variety of resources to help you get started, including documentation, tutorials, and examples.
We are excited to see what you can create with LeoLM!
Komentar