Skip to content

Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦

C

Commits on main

965d175
b1c9d52
1a05695
3059afb
03f6aa8
de27ba0
282b743
5ef694d
aee54ce
cba5f31
0e939fc
b9c679a
127b077
09f1403
39adcc6
f5ec92f
2eb1975
fix(cuda): stop parked issuers before teardown
ZacharyZcR committed 1 month ago
510cfa5
9885e89
6567e76
feat(serve): expose Prometheus scheduler metrics
ZacharyZcR committed 1 month ago
90ff72b
33900c2