llama.cpp Release b10573 Introduces WebP Support via FFmpeg
The llama.cpp project has released build b10573, which adds support for the WebP image format by integrating FFmpeg. This update enhances the library's capability to handle multimodal vision-language models (VLMs) by allowing them to process WebP images directly. The feature was introduced in pull request #27520 and is delivered as part of a routine automated build release.
## BACKGROUND
llama.cpp is an open-source C/C++ library designed for efficient local inference of large language and vision-language models. FFmpeg is a widely used multimedia framework that handles video, audio, and other multimedia files and streams, including image formats like WebP.