CYFRAGOVPL/PLLuM-12B-nc-base-2412
CYFRAGOVPL/PLLuM-12B-nc-base-2412 is a 12 billion parameter large language model from the PLLuM family, developed by a consortium led by Politechnika Wrocławska. Based on Mistral-Nemo-Base-2407, it is specialized in Polish and other Slavic/Baltic languages, with additional English data for broader generalization. This base model is pretrained on up to 150 billion tokens of Polish text and is intended for non-commercial use under a CC-BY-NC-4.0 license. It excels in generating contextually coherent text and serves as a foundation for specialized applications requiring strong Polish language capabilities.
Loading preview...
PLLuM-12B-nc-base-2412: A Polish-Centric LLM
This model is part of the PLLuM family, a series of large language models developed by a consortium led by Politechnika Wrocławska, specialized in Polish and other Slavic/Baltic languages. The PLLuM-12B-nc-base-2412 is a 12 billion parameter base model, built upon the Mistral-Nemo-Base-2407 architecture. It is specifically designed for non-commercial use, licensed under CC-BY-NC-4.0.
Key Capabilities
- Polish Language Specialization: Pretrained on extensive Polish corpora (up to 150 billion tokens), ensuring high proficiency in the language.
- Multilingual Support: Incorporates additional Slavic, Baltic, and English data for broader linguistic generalization.
- High-Quality Training Data: Benefits from large-scale, high-quality text data collection and advanced alignment techniques.
- Foundation Model: Serves as a robust base for further fine-tuning and specialized applications.
Good For
- Research & Development: Ideal for academic or industrial projects requiring a strong command of the Polish language.
- General Language Tasks: Suitable for text generation, summarization, and question answering in Polish.
- Domain-Specific Adaptations: Can be used as a building block for intelligent assistants, particularly in areas like public administration, where domain-aware retrieval is crucial.