This model is a 57.32% smaller version of google/mt5-base optimized for Latin language via vocabulary size reduction using the trimming method.
This trimmed model should perform similarly to the original model with only 32,768 tokens and a much smaller memory footprint. However, it may not perform well for other languages as tokens not commonly used in the selected languages were removed from the vocabulary.

Model Statistics

Metric	Original	Trimmed	Reduction
Vocabulary size	250,112 tokens	32,768 tokens	86.90%
Model size	582,401,280 params	248,560,896 params	57.32%

Mining Dataset Statistics

Number of texts used for mining: 200,000 texts
Dataset: lbourdois/fineweb-2-trimming

Usage

python

from transformers import AutoModel, AutoTokenizer

model_name = "alphaedge-ai/mt5-base-lat-32768"
model = AutoModel.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)

Citations

mT5

@misc{xue2021mt5massivelymultilingualpretrained,
      title={mT5: A massively multilingual pre-trained text-to-text transformer}, 
      author={Linting Xue and Noah Constant and Adam Roberts and Mihir Kale and Rami Al-Rfou and Aditya Siddhant and Aditya Barua and Colin Raffel},
      year={2021},
      eprint={2010.11934},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2010.11934}, 
}

Trimming blog post

@misc{hf_blogpost_trimming,
      title={Introduction to Trimming}, 
      author={Loïck BOURDOIS and Tom AARSEN and Bram VANROY and Christopher AKIKI and Woojun JUNG and Manuel ROMERO and Prithiv SAKTHI},
      year={2026},
      url={https://huggingface.co/blog/lbourdois/introduction-to-trimming}, 
}