Pinned Loading

1. llama-moe-cache llama-moe-cache Public

   Expert cache + predictive prefetch for MoE inference in llama.cpp. A 12GB GPU can run a 120GB model at native speed. At 3% sparsity, a single workstation could theoretically run a 1T parameter mode…

   C++ 10 5

2. GDrop GDrop Public

   An AirDrop-style file transfer tool using Google Drive for cross-platform sharing between mobile devices and computers. GDrop enables quick file sharing from an iPhone to a computer with one-click …

   HTML 2

3. VU-Campusnet VU-Campusnet Public

   A simple script to connect VU-Campusnet on linux if the official script is frustrating for you too

   Shell 1

4. Linux-FaceTime-Migrator Linux-FaceTime-Migrator Public

   Python 1

5. terminal-search terminal-search Public

   a web search cli tool assisted with ai for quick research in between commands

   Shell 2

6. ticket_assistant ticket_assistant Public

   TypeScript 1