Polyglot-OLMo3-7B-SFT-es

This model is a fine-tuned version of allenai/OLMo-3-1025-7B on Spanish synthetic data, using the best teacher-student combination identified in our paper Polyglot Teachers: Evaluating Language Models for Multilingual Synthetic Data Generation.

The training data was generated by google/gemma-3-27b-it and is available in the PolyglotTeachers-SFT-Synth dataset.

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "ljvmiranda921/Polyglot-OLMo3-7B-SFT-es"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

messages = [{"role": "user", "content": "Hola, ¿cómo estás?"}]
inputs = tokenizer.apply_chat_template(messages, return_tensors="pt", add_generation_prompt=True)
outputs = model.generate(inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Acknowledgements

LJVM and AK acknowledge the support of the UKRI Frontier Grant EP/Y031350/1 (EQUATE). This work was performed using joint resources provided by the Cambridge Service for Data Driven Discovery (CSD3) EP/T022159/1 and the Isambard AI National AI Research Resource (AIRR) ST/AIRR/I-A-I/1023, and the Microsoft Research Grant. LJVM would also like to thank Songbo Hu, Chen Cecilia Liu, Millicent Ochieng, and Felermino Ali for helpful and productive discussions on the project.

Citation

@misc{miranda2025polyglotteachers,
  title={{Polyglot Teachers: Evaluating Language Models for Multilingual Synthetic Data Generation}},
  author={Lester James V. Miranda and Ivan Vulić and Anna Korhonen},
  year={2025},
}