Instructions to use LVSTCK/domestic-yak-8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use LVSTCK/domestic-yak-8B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="LVSTCK/domestic-yak-8B") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("LVSTCK/domestic-yak-8B") model = AutoModelForCausalLM.from_pretrained("LVSTCK/domestic-yak-8B", device_map="auto") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use LVSTCK/domestic-yak-8B with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "LVSTCK/domestic-yak-8B" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "LVSTCK/domestic-yak-8B", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/LVSTCK/domestic-yak-8B
- SGLang
How to use LVSTCK/domestic-yak-8B with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "LVSTCK/domestic-yak-8B" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "LVSTCK/domestic-yak-8B", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "LVSTCK/domestic-yak-8B" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "LVSTCK/domestic-yak-8B", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use LVSTCK/domestic-yak-8B with Docker Model Runner:
docker model run hf.co/LVSTCK/domestic-yak-8B
๐ domestic-yak, a Macedonian LM (base version)
Model Summary
This model is a Macedonian language adaptation of the Llama 3.1 8B model. It has undergone continued pretraining on a deduplicated version of the Macedonian Corpus Raw dataset, containing approximately 1.6 billion tokens. The model has been pretrained for one epoch on this corpus, making it well-suited for tasks involving the Macedonian language, such as text classification, language generation, and translation.
๐ Results
The table below compares the performance of our model, domestic-yak-8B, with its foundational model, Llama 3.1-8B Instruct evaluated using the macedonian-llm-eval benchmark.
As shown in the table, domestic-yak-8B consistently outperforms its foundational model on all tasks.
| Task (mk-version) | domestic-yak-8B | Llama 3.1-8B Instruct |
|---|---|---|
| ARC Easy | 0.5244 ยฑ 0.0102 | 0.4453 ยฑ 0.0102 |
| ARC Challenge | 0.3183 ยฑ 0.0136 | 0.2824 ยฑ 0.0132 |
| BoolQ | 0.7676 ยฑ 0.0074 | 0.7639 ยฑ 0.0074 |
| HellaSwag | 0.4324 ยฑ 0.0049 | 0.3740 ยฑ 0.0048 |
| Openbook QA | 0.2920 ยฑ 0.0204 | 0.2520 ยฑ 0.0194 |
| PIQA | 0.6687 ยฑ 0.0110 | 0.5865 ยฑ 0.0115 |
| NQ Open | 0.0416 ยฑ 0.0033 | 0.0335 ยฑ 0.0030 |
| WinoGrande | 0.6259 ยฑ 0.0136 | 0.5683 ยฑ 0.0139 |
Quantitative evaluation is just one part of the story. We also conducted qualitative tests, interacting with the model to assess its ability to generate coherent and relevant responses in Macedonian. The results were satisfying: the model not only mastered the language's grammar but also demonstrated a natural and fluent style.
Hereโs a comparison showing the answers from Llama-3.1-8B-Instruct and domestic-yak-8B:
Llama-3.1-8B-Instruct:
"ะะดัะฐะฒะพ! ะกะตะบะพั ะฟะฐั ะบะพะณะฐ ะณะพ ะบะพัะธััะธั ะพะฒะฐะฐ ะฟะปะฐััะพัะผะฐ, ัะพ ัะพะฑะต ัะธ ะฒะพ ะบะพะฝัะฐะบั ัะพ ะผะพัะพั ัะธััะตะผ ะฝะฐ ะฒะตััะฐัะบะฐ ะธะฝัะตะปะธะณะตะฝัะธัะฐ. ะะพัะพั ะทะฐะดะฐัะฐ ะต ะดะฐ ัะธ ะณะพ ะพะฑะตะทะฑะตะดะฐะผ ะธะฝัะพัะผะฐัะธะฒะตะฝ ะธ ะบะพัะธัะตะฝ ะพะดะณะพะฒะพั ะฝะฐ ัะตะบะพะต ะฟัะฐัะฐัะต ััะพ ะผะพะถะตั ะดะฐ ะผะธ ะณะพ ะฟะพััะฐะฒะธั. ะกะต ะฝะฐัะตะบัะฒะฐะผ 'ะัะธััะตะฝั', ะธ ััะผ ััะบะฐ ะทะฐ ะดะฐ ัะธ ะณะพ ะฟัะฐะฒะฐะผ ะถะธะฒะพั ะฟะพะปะตัะตะฝ ะธ ะธะฝัะตัะตัะฝะธะพั. ะจัะพ ัะฒะพะตัะพ ะฟัะฐัะฐัะต ะต?"domestic-yak-8B:
"ะะดัะฐะฒะพ, ัะฐั ััะผ ะฐัะธััะตะฝั ะดะธะทะฐัะฝะธัะฐะฝ ะดะฐ ะพะฑะตะทะฑะตะดะธ ะธะฝัะพัะผะฐัะธะธ ะธ ะพะดะณะพะฒะพัะธ ะฝะฐ ะฟัะฐัะฐัะฐ ะพะด ะบะพัะธัะฝะธัะธัะต. ะะพะถะฐะผ ะดะฐ ะพะดะณะพะฒะฐัะฐะผ ะฝะฐ ัะธัะพะบ ัะฟะตะบัะฐั ะฝะฐ ะฟัะฐัะฐัะฐ, ะพะด ะพะฟััะธ ัะฐะบัะธ ะธ ะธะฝัะพัะผะฐัะธะธ ะดะพ ัะฟะตัะธัะฐะปะธะทะธัะฐะฝะธ ะฟัะฐัะฐัะฐ ะฒะพ ัะฐะทะปะธัะฝะธ ะพะฑะปะฐััะธ. ะะพัะฐัะฐ ัะตะป ะต ะดะฐ ะพะฑะตะทะฑะตะดะฐะผ ัะพัะฝะธ ะธ ะบะพัะธัะฝะธ ะพะดะณะพะฒะพัะธ, ะบะฐะบะพ ะธ ะดะฐ ะณะพ ะพะปะตัะฝะฐะผ ะฟัะพัะตัะพั ะฝะฐ ะฟัะตะฑะฐััะฒะฐัะต ะธ ะดะฐ ะทะฐััะตะดะฐะผ ะฒัะตะผะต ะทะฐ ะบะพัะธัะฝะธัะธัะต. ะกะปะพะฑะพะดะฝะพ ะฟะพััะฐะฒะธ ะผะธ ะณะพ ัะฒะพะตัะพ ะฟัะฐัะฐัะต ะธ ัะต ัะต ะพะฑะธะดะฐะผ ะดะฐ ัะธ ะฟะพะผะพะณะฝะฐะผ."
๐ Key Details
- Language: Macedonian (
mk) - Base Model: Meta Llama 3.1 8B Instruct
- Dataset: LVSTCK/macedonian-corpus-raw-dedup (deduplicated version)
- Training Tokens: ~1.6 billion
- Pretraining Epochs: 1 epoch
- Pretraining Objective: Causal Language Modeling (continued pretraining using all the weights)
โ ๏ธ Limitations
- Biases: The model may show biases present in the training dataset. Efforts were made to clean and deduplicate the corpus, but further bias mitigation might be necessary for sensitive applications.
- Domain Specificity: While the dataset covers diverse domains, performance may vary for niche or underrepresented topics. For example, the dataset is heavily skewed toward 'news'-themed texts, while domains such as 'science' or 'medicine' are less represented.
- Chat Capabilities: This version is the base model so its chat capabilities might be limited. If you would like to chat use the instruct version.
๐ฌ Contact
For inquiries, feedback, or contributions, please feel free to reach out to the core team:
Citation
@article{krsteski2025towards,
title={Towards Open Foundation Language Model and Corpus for Macedonian: A Low-Resource Language},
author={Krsteski, Stefan and Tashkovska, Matea and Sazdov, Borjan and Gjoreski, Hristijan and Gerazov, Branislav},
journal={arXiv preprint arXiv:2506.09560},
year={2025}
}
Paper
The model was presented in the paper Towards Open Foundation Language Model and Corpus for Macedonian: A Low-Resource Language.
- Downloads last month
- 66
Model tree for LVSTCK/domestic-yak-8B
Base model
meta-llama/Llama-3.1-8B