Community asks for GLM Air model and GGUF quants

Reddit users request a smaller, locally-runnable GLM Air model, noting that GLM 5.1 is a powerful coder but too large for local use. They also call for GGUF quantizations to enable local inference.