setu4993/convert-labse-tf-pt

Convert LaBSE model from TF Hub to PyTorch.

41
/ 100
Emerging

This project helps machine learning engineers and researchers convert powerful LaBSE and LEALLA language models from Google's TensorFlow format to PyTorch. You input a TensorFlow version of these models, and it outputs a PyTorch-compatible model, ready for use in different machine learning frameworks and deployments. This is for machine learning practitioners working with multilingual text who need flexibility across deep learning ecosystems.

Available on PyPI.

Use this if you need to use Google's LaBSE or LEALLA multilingual embedding models within a PyTorch environment or wish to upload them to the Hugging Face Model Hub in PyTorch, TensorFlow, or Flax formats.

Not ideal if you are not a machine learning engineer or do not need to work directly with deep learning model formats.

natural-language-processing machine-learning-engineering multilingual-embeddings deep-learning-frameworks model-conversion
Maintenance 10 / 25
Adoption 6 / 25
Maturity 25 / 25
Community 0 / 25

How are scores calculated?

Stars

16

Forks

Language

Python

License

MIT

Last pushed

Jan 15, 2026

Commits (30d)

0

Dependencies

6

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/embeddings/setu4993/convert-labse-tf-pt"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.