Tag
This pull request adds video input support to llama.cpp, enabling multimodal models to process video data via the new mtmd component.