Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦
C
release: build and copy the engine under its real name (glm -> colibri)
JustVuggb3b6e8d4de12
diff --git a/.github/workflows/release.yml b/.github/workflows/release.ymlindex a556b80b..638dc7f0 100644--- a/.github/workflows/release.yml+++ b/.github/workflows/release.yml@@ -52,8 +52,8 @@ jobs:- name: Build enginerun: |cd c- make glm ${{ matrix.make_args }}- ls -lh glm${{ matrix.ext }}+ make colibri ${{ matrix.make_args }}+ ls -lh colibri${{ matrix.ext }}- name: Packagerun: |@@ -64,7 +64,7 @@ jobs:# colibri-<tag>-<platform>.exe made every downloaded archive fail with# "engine is not built. Run: coli build" -- with the engine sitting right there.# The version lives in the ARCHIVE name, which is where a downloader looks for it.- cp c/glm${{ matrix.ext }} dist/colibri${{ matrix.ext }}+ cp c/colibri${{ matrix.ext }} dist/colibri${{ matrix.ext }}cp c/coli dist/cp c/version.py dist/cp c/openai_server.py dist/