はてなブックマーク - GitHub - JustVugg/colibri: Run GLM-5.2 (744B MoE) on a 25GB-RAM consumer machine — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦 https://b.hatena.ne.jp/entry/s/github.com/JustVugg/colibri A 744B Mixture-of-Experts model activates only ~40B parameters per token — and only ~11 GB of those change from token to token (the routed experts). So: the dense part (attention, shared experts, e... tasukuchan https://b.hatena.ne.jp/tasukuchan/20260710#bookmark-4789955543702508674 2026-07-10T05:54:36Z barlog https://b.hatena.ne.jp/barlog/20260710#bookmark-4789955543702508674 2026-07-10T04:54:26Z