ai4sd/multiscale-byte-lm

A hierarchical LM that scales to training on context windows of +5M tokens

31
/ 100
Emerging

This is a developer tool that provides a hierarchical language model architecture capable of processing extremely long sequences of text, up to millions of tokens. It takes in raw byte-level text data and outputs predictions or loss values, which are then used to train and evaluate large language models. This tool is designed for machine learning engineers and researchers who are building or experimenting with advanced language models.

Use this if you are a machine learning engineer or researcher focused on developing new language models and need an architecture that can handle very long text contexts efficiently.

Not ideal if you are an end-user looking for a pre-trained language model or a high-level API for natural language processing tasks.

Machine Learning Engineering Natural Language Processing Large Language Models Deep Learning Research Model Architecture
No Package No Dependents
Maintenance 10 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 0 / 25

How are scores calculated?

Stars

9

Forks

Language

Python

License

MIT

Last pushed

Feb 17, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/ai4sd/multiscale-byte-lm"

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