Jul 12, 2026
•
10 min read
colibrì: How a Single C File Runs a 744B-Parameter MoE Model on 25 GB of RAM
colibrì is a single-file, dependency-free C engine that runs GLM-5.2, a 744B-parameter MoE model, on consumer hardware with as little as 25 GB of RAM by streaming experts straight from NVMe disk.
Ad
Admin