Small language model

Microsoft today announced Phi-4, a 14B-parameter "state-of-the-art" small language model (SLM) that specializes in complex reasoning in areas such as mathematics, in addition to traditional language processing. Phi-4 is the latest addition to the Phi family of small language models, and officials say it demonstrates Microsoft's commitment to continue exploring the boundaries of SLMs. Possibilities. Officials say Phi-4 outperforms comparable and larger models in mathematical reasoning thanks to a number of technological advances, including the use of high-quality synthetic datasets, carefully selected high-quality organic data, and post-training innovations. Its mathematical ...

Recently, Hugging Face launched a new AI tool - SmolLM. This is a series of high-performance small language models with parameters ranging from 135M to 1.7B, designed specifically for various devices and applications. Imagine that these small models can run efficiently on mobile phones and laptops, which is really cool! The characteristics of SmolLM models are small and powerful. They can still perform well with less computing resources and help users protect privacy. Hugging Face uses a tool called SmolLM-Corp when training these models...