Live page · Day archive

Compress any LLM 3x in one click,near-lossless,in under 50 MB RAM

Read the original at Hacker News

Summary

SPECTRA compresses large language models three times with near-lossless accuracy using under 50 megabytes of RAM and achieves 82.53 tokens per second on 4GB consumer GPUs.

Carried by: Hacker News. First seen: .