llama.cpp Release b11563 Updates GGML Symbol Visibility
Llama.cpp has released version b11563, which changes the default visibility of GGML symbols to hidden. This routine update prevents a small number of internal symbols from unintentionally leaking through the public GGML API. While this is a minor internal code cleanup rather than a major feature addition, it improves library hygiene and encapsulation. Maintaining strict symbol visibility helps prevent namespace pollution and potential conflicts for software linking against the GGML library. The change was introduced via pull request #30014 following community discussion in issue #29997. It specifically restricts internal components from being exposed outside of their intended scope.
## BACKGROUND
Llama.cpp is an open-source software library used to perform inference on large language models like Meta's Llama on local hardware. It is built on top of GGML, a general-purpose tensor library designed for machine learning workloads.