Skip to content

armgit / Edge0

Public Imported from GitHub
0

Run 35B MoE models in ~2.5 GB of RAM. Weights stream from SSD; a trained router predicts loads a step ahead. Mac, iPhone, Android, Windows supported

Python

Commits on main

08fcb46
fix(scripts): upload README.md in upload_hf.py
Demetrios Agourakis committed 1 month ago
62295c2
3472045
5b65e6f
9a83488
docs: drop the peak-memory footnote boilerplate
linyubupa committed 2 months ago
e4cd73c
docs: drop the peak-memory footnote boilerplate
linyubupa committed 2 months ago
dd633cd
docs: drop the peak-memory footnote boilerplate
linyubupa committed 2 months ago
2982774
56c8b79
b909856
48bd224
b413a22
63d6efb
6cc13ef
fix(streaming): give the hot stack its zero overflow row
Demetrios Agourakis committed 2 months ago
1e6f8c6