A pull request #27773 to ggml-org/llama.cpp adds support for GLM-5.3‑Flash (GLM5‑Next), enabling the model to be run on home computers. The contribution by timkhronos was highlighted in a Reddit post on r/LocalLLaMA.