CYFRAGOVPL/Llama-PLLuM-70B-instruct-2512
The CYFRAGOVPL/Llama-PLLuM-70B-instruct-2512 is a 70 billion parameter instruction-tuned large language model developed by the PLLuM consortium and HIVE AI, specialized in Polish with additional English data. Built upon the Llama-3.1-70B architecture, it features extensive Polish data collection, including a unique "organic instructions" dataset and the first Polish-language preference corpus. This model excels in general language tasks and is particularly effective for domain-specific applications within Polish public administration, achieving state-of-the-art results in Polish-language benchmarks.
Loading preview...
Overview
CYFRAGOVPL/Llama-PLLuM-70B-instruct-2512 is a 70 billion parameter instruction-tuned model from the PLLuM family, developed by the PLLuM consortium and continued by HIVE AI. This model is built on the Llama-3.1-70B base and is highly specialized for the Polish language, incorporating significant English data for broader generalization. Its development involved extensive data collection, rigorous cleaning, and advanced alignment techniques, including instruction tuning and preference learning.
Key Capabilities
- Polish Language Specialization: Developed with large-scale, high-quality Polish text corpora, ensuring strong performance in Polish-language tasks.
- Organic Instruction Dataset: Utilizes a unique, manually created Polish "organic instructions" dataset, designed to cover subtle aspects of human-model interactions and mitigate negative linguistic transfer.
- Polish Preference Corpus: Features the first Polish-language preference corpus, manually assessed for correctness, balance, and safety, particularly for sensitive topics.
- Robust Evaluation: Achieves top scores on custom benchmarks relevant to Polish public administration and state-of-the-art results in broader Polish-language tasks.
- Retrieval Augmented Generation (RAG): Post-trained to perform effectively in RAG settings, capable of answering questions based on provided documents and citing sources.
Good For
- General Language Tasks: Ideal for text generation, summarization, extraction, and question answering in Polish.
- Domain-Specific Assistants: Especially effective for applications within Polish public administration, legal, or bureaucratic contexts requiring domain-aware retrieval.
- Research & Development: Serves as a strong foundation for AI applications demanding a high command of the Polish language.