Search AI Products and News

Microsoft officially announced the release of a new device-side small language model (SLM), Mu, specifically designed for the settings application in Windows 11 to create an intelligent AI agent. This marks a significant breakthrough in localized AI technology within operating system interactions. As a compact and efficient model, Mu runs through deep optimization on the neural processing unit (NPU), providing users with a low-latency, high-privacy natural language interaction experience. AIbase has compiled the highlights and industry impact of Mu, taking you on a journey to explore its features. Mu: Designed Specifically for Windows Settings

This morning, Microsoft officially announced its latest innovative small-parameter model, Mu. This model has only 330 million parameters, yet it can match the performance of Microsoft's previously released Phi-3.5-mini, while being just one-tenth the size of Phi-3.5-mini. More notably, Mu can achieve a response speed of over 100 tokens per second on offline NPU laptops, which is a rare breakthrough in the field of small-parameter models. A major highlight of the Mu model is its support for setting intelligent features in Windows.

The French AI lab Mistral recently announced the official release of its first inference model series -- Magistral. The series includes two versions: Magistral Small and Magistral Medium, aimed at enhancing logical reasoning capabilities in disciplines such as mathematics and physics. Mistral stated that the Magistral series models solve problems step by step to improve the consistency and reliability of results.

In the field of AI art generation, there has long been a general understanding that generating high-quality images and videos requires larger models, more parameters, and stronger computing power. However, the research team from Hong Kong University of Science and Technology and Kuaishou Technology recently proposed the EvoSearch (evolutionary search) technology, which is completely overturning this traditional notion. The most impressive performance of this technology is: after using EvoSearch, the generation quality of the Stable Diffusion 2.1 model with only 865M parameters has astonishing results.

X platform recently announced a major update to its recommendation algorithm, deeply integrating self-developed Grok AI technology, aiming to optimize user content experience and significantly increase the visibility opportunities for small account content. This adjustment is considered by the industry as an important innovation in the traditional social media content distribution mechanism. The core upgrade of the algorithm: quality first, scale second. The core of the latest algorithm update focuses on providing more visibility to small accounts, and the platform will now highlight content from lesser-known creators, creating a more equitable competitive environment. The new algorithm prioritizes推送 through Grok AI technology.

X Platform announced a major update to its recommender algorithm, deeply integrating Grok AI technology, with the aim of optimizing user content experience. This adjustment will significantly increase the exposure opportunities for high-quality content from small accounts in the recommendation feed, while also optimizing the sorting mechanism for comment replies. The new algorithm uses Grok AI to prioritize high-quality content, especially viral posts from small accounts with fewer followers. Users will see more diverse and creative posts in their recommended feeds, rather than being limited to content from big-name influencers. This move will provide more display opportunities for newcomers and creators.

IBM recently announced the establishment of a new watsonx AI Labs innovation center in New York City, aiming to accelerate the development of AI technology and promote its large-scale application. The laboratory is located at IBM's office in Manhattan, designed to connect AI developers and startups with IBM's research and engineering networks, supporting the development of enterprise-level intelligent AI systems. The watsonx AI Labs will focus on solving everyday business challenges, including customer service, supply chain optimization, etc.

JD Industrial officially released the industry's first supply chain-centric industrial large model -- Joy industrial in Shanghai. This innovative product aims to improve the operational efficiency of upstream and downstream enterprises in the supply chain through advanced artificial intelligence technology. Joy industrial provides various AI agents for JD Industrial and its upstream suppliers, including demand agent, operation agent, customs agent, etc., and also launches AI products such as product expert and integration expert for downstream enterprises in the supply chain.

With the ongoing impact of Trump-era tariffs, businesses face increasing geopolitical risks. Making effective decisions in a rapidly changing environment presents a significant challenge for corporations, non-profits, and consulting firms. London-based startup Hence AI recently launched Hence Global, an AI software product designed to help organizations monitor and manage these risks. Hence Global offers two primary functions: Firstly, it provides risk tracking for companies of all sizes.

Xie Shaofeng, the chief engineer of the Ministry of Industry and Information Technology (MIIT), stated that over 400 national-level specialized and innovative small and giant enterprises have been cultivated in the artificial intelligence field. The next step is to guide patient capital to increase support, accelerate the cultivation of a group of industry-leading enterprises and specialized and innovative SMEs. This includes building an open-source AI community, leveraging the role of the AI Standardization Technical Committee, and accelerating the development of key and urgently needed standards.

New York-based AI startup 1Fort announced a $7.5 million seed funding round to improve the process of small businesses obtaining commercial insurance through its AI-powered platform. The company, which saw 200% revenue growth in 2024, aims to automate cumbersome and outdated manual processes, helping millions of underserved small businesses address insurance gaps. The funding round was led by Bonfire Ventures, with participation from Draper Associates.

Mistral has released Mistral Small 3.1, an open-source large language model boasting a 128K context window and superior parameter efficiency compared to GPT-4o Mini.