创业邦·2025-09-20 04:39Group 1 - Xiaomi has officially open-sourced its first native end-to-end voice model, Xiaomi-MiMo-Audio, which is based on an innovative pre-training architecture and over a billion hours of training data, achieving few-shot generalization based on ICL in the voice domain and observing significant "emergent" behavior during pre-training [2] - Ashish Kumar, head of Tesla's Optimus AI team, has announced his departure to join Meta as a research scientist. During his tenure at Tesla, he focused on scalable AI methods and enhanced robot dexterity through reinforcement learning [2] - Google is integrating its Gemini AI model into the Chrome browser, allowing users to request explanations of web pages, consolidate information from multiple tabs, and restore previously closed sites. This integration follows a court ruling that Google does not need to divest Chrome [2] - Tencent has launched a one-stop work platform called "混元3D Studio," aimed at 3D designers and game developers. The platform utilizes AI technology to streamline the entire 3D production process, reducing production cycles from days to minutes [2]