The ggml-org team has released llama.cpp b12422, highlighting four added Jinja features. Jinja is a template language used to format prompts for models.
The release page lists downloads for macOS, iOS and several Linux configurations, including CPU and Vulkan builds. It does not name the four features in the supplied release notes.
Why it matters
People running llama.cpp on the listed platforms can download a new build, but the brief release note does not explain which Jinja workflows the four additions affect.