llama.cpp Release b12479 Adds Support for Minimax-Music3
Llama.cpp release b12479 has been officially launched, introducing support for the Minimax-Music3 multimodal model alongside standard prebuilt binaries for various operating systems. This update expands the range of architectures supported by the library, allowing users to run advanced multimodal models locally across a variety of hardware backends. The release includes prebuilt binaries for macOS, Linux, Android, and Windows, supporting multiple backends such as CPU, CUDA, Vulkan, and Snapdragon NPUs.
## BACKGROUND
llama.cpp is an open-source software library created by Georgi Gerganov that performs inference on large language models using C/C++. It serves as a core infrastructure for running local AI models efficiently on consumer hardware.