metadata

license: apache-2.0
language:
  - hu
  - en
  - zh
tags:
  - puli

PULI Trio Q 7B base (7.62B billion parameter)

Trained with LLaMA-Factory github
The Qwen2.5 7B Instruct model were continual pretrained on Hungarian dataset

Dataset for continued pretraining

Hungarian (8.08 billion words): documents (763K) that exceed 5000 words in length + Hungarian Wikipedia
English: Long Context QA (2 billion words), BookSum (78 million words)
Chinese (3 billion Chinese characters): Wudao
The training was completed using a Hungarian-only dataset:
- 626 million Hungarian words (1 epoch): Hungarian Wikipedia + News articles

Limitations

max_seq_length = 32 768

Usage with pipeline

from transformers import pipeline, Qwen2ForCausalLM, AutoTokenizer

model = Qwen2ForCausalLM.from_pretrained("NYTK/PULI-Trio-Q")
tokenizer = AutoTokenizer.from_pretrained("NYTK/PULI-Trio-Q")
prompt = "Elmesélek egy történetet a nyelvtechnológiáról."
generator = pipeline(task="text-generation", model=model, tokenizer=tokenizer, device=0)

print(generator(prompt, max_new_tokens=30)[0]["generated_text"])

Citation

If you use this model, please cite the following paper:

@inproceedings {yang-llumix-llama,
    title = {PULI Chat: Our First Hungarian Conversational Model},
    booktitle = {International Conference on Formal Methods and Foundations of Artificial Intelligence},
    year = {2025},
    publisher = {Eszterházy Károly Catholic University},
    address = {Eger, Hungary},
    author = {Yang, Zijian Győző and Bánfi, Ágnes and Dodé, Réka and Ferenczi, Gergő and Földesi, Flóra and Hatvani, Péter and Héja, Enikő and Lengyel, Mariann and  Madarász, Gábor and Osváth, Mátyás and Sárossy, Bence and Varga, Kristóf and Váradi, Tamás and Prószéky, Gábor and Ligeti-Nagy, Noémi},
    pages = {1--3},
    pubstate={accepted abstract},
    url ={https://uni-eszterhazy.hu/api/media/file/7f9158bd443acc29dbd2a211971fe8677768257c}
}