library
JustVugg/colibri
It can run a massive model on a consumer PC, but sparse docs and steep setup limit its practicality.
4.9Overall
Utility5
Onboarding4
Craft4
Niche fit7
Longevity4
Five dimensions scored independently; overall is a weighted average. Scores are only comparable within this same rubric.
Good for
Researchers or hobbyists who want to experiment with massive LLMs locally and are comfortable building C code
Not for
Teams needing a stable, production‑ready solution or lacking C compilation expertise
Project description
Run GLM-5.2 (744B MoE) on a 25GB-RAM consumer machine — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦
Alternatives
llama.cppggmltext-generation-webui