robtacconelli/Nacrith-GPU

Nacrith — Lossless text compression via ensemble neural arithmetic coding. Combines SmolLM2-135M language model with context mixing, adaptive prediction, and high-precision CDF coding. 3.1x better than gzip, and outperforming CMIX, ts_zip, LLMZip and FineZip. Fully lossless. GPU accelerated.

30
/ 100
Emerging

Nacrith GPU helps you dramatically shrink the size of your text data without losing any information. It takes plain text, uses a small AI model to 'understand' the language, and outputs a much smaller compressed file. This tool is perfect for anyone managing large volumes of text, like researchers archiving documents or data professionals storing logs.

Use this if you need to store or transmit large text files and want the absolute best compression ratios, significantly outperforming standard tools like gzip or zip.

Not ideal if your primary concern is compression speed over file size, or if you don't have access to a GPU.

data-archiving text-storage document-management log-file-compression data-transmission
No Package No Dependents
Maintenance 13 / 25
Adoption 6 / 25
Maturity 11 / 25
Community 0 / 25

How are scores calculated?

Stars

17

Forks

Language

Python

License

Apache-2.0

Last pushed

Mar 21, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/robtacconelli/Nacrith-GPU"

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