Core Insights - Luo Fuli has officially joined Xiaomi as the head of the MiMo team, marking a significant step in the company's AI ambitions [1][3] - The evolution of intelligence is transitioning from the language domain to the physical world, aiming to unlock multi-modal spatial intelligence, which is crucial for achieving true Artificial General Intelligence (AGI) [4] Timeline and Background - Rumors about Luo Fuli joining Xiaomi surfaced last year, with reports indicating she was recruited by Lei Jun with a salary of tens of millions [5][10] - Key dates include the launch of DeepSeek-V3 on December 25, followed by media reports of Xiaomi assembling a GPU cluster [6][7] - On December 31, Lei Jun publicly shared Xiaomi's ambitions in AI during a New Year's live stream [8] Luo Fuli's Profile - Luo Fuli holds a Bachelor's degree in Computer Science from Beijing Normal University and a Master's in Computational Linguistics from Peking University, with numerous publications in top NLP conferences [15] - She has worked at Alibaba's DAMO Academy and DeepSeek, contributing to the development of various deep learning models [17] - Her academic work has garnered over 11,000 citations, with approximately 8,000 citations added in the past year alone [18] Xiaomi's AI Strategy - The MiMo project is central to Xiaomi's efforts in advancing large model research, with a focus on spatial intelligence [24] - Spatial intelligence aims to bridge the gap between information AI and physical AI, aligning with Xiaomi's ecosystem of people, vehicles, and homes [26]

Core Viewpoint - The article highlights the appointment of Luo Fuli as the head of Xiaomi's MiMo team, focusing on advancing spatial intelligence as a key step towards achieving Artificial General Intelligence (AGI) [1][3][23]. Group 1: Appointment and Background - Luo Fuli officially announced her role at Xiaomi on November 12, leading the MiMo team [1]. - She previously worked at DeepSeek and was reportedly recruited by Lei Jun with a salary of tens of millions [4][7]. - Luo has a strong academic background, with over 11,000 citations for her research papers, indicating her prominence in the field [17][18]. Group 2: MiMo Team and Objectives - The MiMo team aims to unlock multi-modal spatial intelligence, which includes perception, reasoning, generation, and action capabilities [4][23]. - Luo's vision aligns with the broader goal of integrating information AI with physical AI, creating a seamless connection between the digital and physical worlds [25]. Group 3: Industry Context and Implications - The concept of spatial intelligence has gained attention, with AI experts like Fei-Fei Li discussing its significance for embodied intelligence and AGI [24]. - Xiaomi's focus on spatial intelligence is seen as a strategic move, leveraging its ecosystem that includes people, vehicles, and homes [25].

Investment Rating - The report indicates that Alibaba's Qwen currently ranks at the top of the open-source model rankings, with expectations for continued leadership in model capability and ecosystem monetization [2]. Core Insights - The report highlights a surge in domestic open-source models, with significant releases from Alibaba, Xiaomi, and DeepSeek, showcasing advancements in large language models (LLMs) [1][8]. - Alibaba's Qwen-3 series demonstrates substantial performance improvements, achieving 10-30% accuracy gains on various benchmarks and enhancing inference speed by 20-40% [9][12]. - Xiaomi's MiMo model, with 7 billion parameters, excels in reasoning and code generation tasks, outperforming larger proprietary models through innovative training strategies [10][12]. - DeepSeek's Prover-V2-671B model shows strong performance in formal logic reasoning, indicating a strategic focus on specialized AI applications [11][12]. - The report anticipates that as more domestic models are released, the industry may face challenges related to homogenization and competition, pushing for more customized solutions in vertical industries [5]. Summary by Sections Alibaba Qwen-3 - The Qwen-3 series includes models ranging from 1.5 billion to 72 billion parameters, designed for various inference needs, with notable performance enhancements over previous generations [9]. - Deployment costs are significantly lower, requiring only 4 H20 GPUs for full-capacity operation, which is advantageous compared to similar models from OpenAI and Grok [2][12]. Xiaomi MiMo - MiMo's training involved 25 trillion tokens and innovative mechanisms to improve training efficiency, achieving a 2.29x increase in training speed and a 1.96x acceleration in verification processes [10]. DeepSeek-Prover-V2-671B - This model excels in mathematical theorem proving, particularly in formal logic, and serves as a precursor to DeepSeek's upcoming models, reflecting the company's commitment to advancing AI capabilities [11]. Industry Trends - The report suggests that the next phase for open-source models will involve customization based on user data and feedback, aiming to establish long-term barriers and user loyalty in specific industries [5].