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

Merge pull request #833 from JustVugg/fix/serve-broken-pipe
Vincenzo Fornaro committed 3 months ago
ecc94ee
serve: a client hanging up is not a server error
JustVugg committed 3 months ago
f5fcbc1
Merge pull request #832 from JustVugg/feat/banner-model
Vincenzo Fornaro committed 3 months ago
cc65879
coli: the banner said GLM-5.2 over every model
JustVugg committed 3 months ago
e2fe0fe
Merge pull request #831 from JustVugg/feat/v4-kv-prefix
Vincenzo Fornaro committed 3 months ago
ecd4cc8
9d096dc
Merge dev into docs/environment-coverage
JustVugg committed 3 months ago
9038201
Merge pull request #808 from ZacharyZcR/ci/efficiency-gate
Vincenzo Fornaro committed 3 months ago
3ad9672
Merge pull request #803 from ZacharyZcR/feat/docker-ci
Vincenzo Fornaro committed 3 months ago
95176c8
Merge pull request #817 from kreuzzelg/cuda-fp8-e4m3
Vincenzo Fornaro committed 3 months ago
dc0c979
Merge pull request #757 from rgbkrk/t2-overlap
Vincenzo Fornaro committed 3 months ago
8df3fbc
18d7c3a
9788085
Merge branch 'dev' into feature/windows-hip-native
Kenneth-Javier Ortigoza committed 3 months ago
aa60f7b
6576069
Merge pull request #165 from whale-agent-lab/dev
JustVugg committed 3 months ago
c024a46
Merge pull request #806 from ZacharyZcR/perf/omp-grain-guard
Vincenzo Fornaro committed 3 months ago
83b06ac
1e12047
a4d7b90