Run 744B‑parameter GLM‑5.2 on a 25 GB PC with a single C file
Colibri lets you run Meta’s 744‑billion‑parameter GLM‑5.2 MoE model on a consumer machine with ~25 GB RAM using only a 2‑k‑line C engine. It streams experts from disk, needs no Python or GPU, and delivers near‑full‑model quality in under a minute per token.