
Hugging Face has released EMO, a mixture-of-experts model designed to achieve modularity without relying on predefined human biases. EMO allows users to utilize only a small subset of its experts for specific tasks while maintaining near full-model performance, addressing the inefficiencies of large language models. This model supports selective expert use, reducing computational costs and enhancing flexibility. EMO's architecture encourages coherent expert grouping, making it a versatile tool for various applications.
Read originalThe b9075 release of llama.cpp brings a notable improvement for CUDA users by integrating the snake activation function into a single elementwise kernel. This enhancement is particularly advantageous for audio decoders like BigVGAN and Vocos, which previously depended on a more complex five-operation sequence. By streamlining these operations, the update promises better performance and efficiency across data types such as F32, F16, and BF16. This development reflects llama.cpp's ongoing focus on refining its CUDA capabilities, making it a more compelling option for developers dealing with complex activation functions.
The latest b9076 release of llama.cpp quietly expands its platform support, making it more versatile for developers across various systems. Notably, it now exposes child model information from the router's /v1/models endpoint, enhancing transparency and control for users. The update includes support for macOS Apple Silicon with KleidiAI enabled, as well as expanded compatibility with Ubuntu and Windows systems, including Vulkan and ROCm 7.2. This release doesn't introduce new models but strengthens llama.cpp's position as a flexible inference runtime across diverse hardware configurations.