Drop-in KV cache compressor for local LLM inference - Run 70B models on 8GB RAM
A quick, honest read on the project and why it belongs in your public portfolio.
Drop-in KV cache compressor for local LLM inference - Run 70B models on 8GB RAM
A short narrative that makes the page feel like a real portfolio entry, not a bare repo link.
Drop-in KV cache compressor for local LLM inference - Run 70B models on 8GB RAM
This page gives the project a clean landing spot for GitHub Pages, search, and sharing. It also keeps the repository discoverable alongside your papers.
This repository is an original project, and is surfaced as a primary portfolio item.
Direct links between the project and its publication track.