SenseTime's performance report for the first half of 2024 showcases strong capabilities in the AI sector, particularly with a 256% year-over-year growth in its generative AI business, which for the first time surpassed traditional AI, accounting for 60% of total revenue. Total revenue reached 1.74 billion yuan, a 21% increase year-over-year, although the net loss was 2.47 billion yuan, with the loss margin narrowed by 21% compared to the previous year. The impressive performance of the generative AI business marks a strategic shift in the company's focus, with significant technical strength as its large model 'Ri Ri Xin' outperformed GPT-4Tu in several evaluations.

SenseTime's 2024 mid-year earnings report reveals significant growth in the first half of the year, with revenue reaching 1.74 billion RMB, a year-on-year growth rate of 21%. The generative AI business has become a major source of revenue, increasing by 256% and accounting for 60.4% of total revenue. Xu Li, the company's executive chairman and CEO, stated that the strategic shift towards generative AI has achieved initial results, coinciding with a golden period for AI development. Strong performance is mainly attributed to robust demand in the Chinese market for large models and intelligent computing services, along with the growth of 'SenseNova'.

In 2023, the scale of China's artificial intelligence software market reached 37.74 billion yuan, a year-on-year increase of 26.2%. The computer vision market reached 10.11 billion yuan, led by SenseTime. In the AI voice and semantic market, Baidu Smart Cloud ranked first for the first time, followed closely by iFlytek and Alibaba Cloud. The market size of machine learning development platforms was 3.51 billion yuan, with Fourth Paradigm and Huawei Cloud tied for first place, focusing on large model technology to support the development of generative AI applications.

In 2023, the market size for large model platforms and related applications in China reached 1.765 billion RMB, with the market still in an early stage of investment and observation. Baidu Intelligent Cloud ranked first with a market share of 19.9%, followed closely by SenseTime, while ZhiPu AI stood out as a leading startup in third place. Lu Yanxia, IDC's Research Director in China, pointed out that the market will undergo several rounds of dramatic changes in the next 2-3 years, and continuous investment in foundational large model research and enhancing base model capabilities is key for companies to maintain their competitiveness. Emerging fields such as multimodal large models and scientific computing are also on the rise.

The medical large model 'Da Yi' developed by SenseTime Medical has successfully passed the algorithm filing by the National Cyberspace Administration. This filing marks the authoritative recognition of 'Da Yi's' compliance and safety, and highlights a key progress in its technological application. The model is based on the large language model 'Ri Ri Xin · Shang Liang' with a scale of hundreds of billions of parameters.

'Instant Sketch Fun' is a WeChat mini program for creative photography and selfie collaboration, developed based on SenseTime's 'Daily New · Instant Sketch' text-to-image large model. Since its launch in trial operation, the mini program has quickly gained widespread popularity among users, with the number of registered users and daily active users continuously rising. On the 9th day of trial operation, daily active users exceeded 520,000, with total users surpassing 1.46 million, daily page views reaching 41.17 million, and over 100 million network requests.

SenseTime has collaborated with Shanghai Oriental Media Technology Co., Ltd. (SMT) to apply AI 3D motion capture technology and AR special effects rendering engine for the first time in the archery events at the Paris Olympics, jointly creating the 'InnoMotion AI Event Broadcasting Upgrade Solution.' This technology enhances the collection and visualization of archery event data through 3D motion capture and AR effects, providing viewers with a more in-depth and intuitive viewing experience.

SenseTime Technology has collaborated with the Construction Society to launch the 'Regulations and Standards Knowledge Model for the Construction Sector', based on the 'SenseNova5.5' framework. This model integrates industry standards, laws, design, and process specifications, aiming to provide intelligent and professional knowledge support for various stages of construction. The model has been applied in the knowledge service platform for regulations and standards in housing and urban-rural construction, featuring functions such as intelligent Q&A, integration, duplication check, and comparison to enhance knowledge retrieval efficiency. Trained on high-quality construction knowledge data of 600 million tokens, the model covers 8,000 industry-specific books.

商汤科技与国家篮球队战略合作，通过其“日日新SenseNova5.5”大模型技术开发AI篮球产品，旨在深度整合AI与篮球、体育产业。商汤成为篮球队的人工智能战略伙伴，共同创建运动表现分析平台，利用AI技术提升运动员训练效率和竞技水平。技术通过三维模型和动作捕捉算法实时分析球员和篮球运动，生成教练决策支持，优化训练策略。在三人篮球国家队训练中应用，助力提升竞技水平及国际竞争力。AI产品不仅在竞技体育，还适用于青训体育与大众体育领域，预示智能技术成为提升运动员表现的强大驱动力。此次合作标志着智能与体育的全面深入融合，为体育强国建设贡献力量。

SenseTime Technology, in collaboration with Thailand's DTGO Group and Quinnnova, jointly launched the Thai Language Large Model (DTLM) named "Dongfeng". This is the world's first AI large language model that can operate efficiently in Thai, Chinese, and English. The model combines SenseTime's base model and computational advantages with DTGO's deep understanding of Thai language and culture, aiming to provide localized generative AI experiences, including text comprehension and natural, fluent r

At the 2024 World Artificial Intelligence Conference (WAIC), SenseTime Technology launched a controllable character video generation large model named Vimi. The Vimi model utilizes SenseTime's advanced large model technology to generate character videos consistent with the target action from a single photo, achieving precise facial and limb control. The model supports various driving methods, including videos, animations, sounds, and text. It boasts years of accumulated facial tracking technolog

At the 2024 World Artificial Intelligence Conference, SenseTime Technology released the first domestic "What You See Is What You Get" model, named "Ri Ri Xin 5o". This model offers an interactive experience comparable to GPT-4o, achieving real-time streaming of multi-modal interactions. By integrating cross-modal information such as sound, text, images, and video, it can understand and respond in real-time. For example, it can recognize the name tags worn by staff, determine the venue location,

SenseTime officially launched its groundbreaking "Vimi" controllable character video generation large model.Vimi, as a product developed by SenseTime based on its powerful "Ri Ri Xing" large model capabilities, has achieved unprecedented flexibility in video generation. It can flexibly accept diverse inputs such as motion videos, exquisite animations, rich sound materials, and even textual descriptions as driving elements. It accurately controls and transforms character images to generate videos

Recently, an AI digital human of CCTV journalist Wang Bingbing created by SenseTime's "RuYing AI Video Generation Platform" made its debut at the "2024 China AI Summit". AI Bingbing's performance on stage was both professional and natural, thanks to SenseTime's "RiRiXin SenseNova" large model technology, which allows her to fluently use multiple languages such as Turkish, Russian, and Spanish, showcasing the remarkable achievements of AI technology in language capabilities.SenseTime's RuYing pla

SenseTime has launched the AI office assistant 'Office Raccoon', developed based on large model capabilities, officially introducing the WeChat Mini Program version 'Raccoon Smart Assistant' today. Users can directly insert various file formats (such as xls/xlsx/csv/txt/json) in the WeChat Mini Program and can extract key points, generate charts, and output results in a streaming manner through simple conversation.

Tang Xiaowu, the founder of SenseTime and an artificial intelligence scientist, passed away unexpectedly due to illness on December 15 at the age of 55. Professor Tang was a pioneer in computer vision in China, achieving a world-leading position in facial recognition technology and natural language processing. The multimedia lab he established at the Chinese University of Hong Kong has nurtured a group of young talents in the CV field, having a profound impact on the Chinese CV community.

SenseTime has launched 'Code Raccoon', a smart programming assistant based on a large language model. Code Raccoon supports over 30 mainstream programming languages and popular IDEs, enhancing developer productivity. It features powerful code generation and Chinese language understanding capabilities. Code Raccoon's functions include code generation, comment generation, code correction, and support for user-defined requirements. Its launch is set to significantly improve the programming efficiency of Chinese developers and help address programming challenges.

SenseTime has launched an upgraded version of its medical health large language model 'Da Yi', which supports personalized functions, high-quality training, and low-threshold deployment. It includes 13 preset scenarios to meet customization and personalization needs, collaborating with multiple healthcare institutions to realize practical applications in specific scenarios, providing strong momentum for the medical health industry chain.

InternLM-123B has shown outstanding performance in recent competitions among top large models, ranking first in 12 metrics and surpassing some performance levels of GPT-4. SenseTime's large model has iterated rapidly, upgrading from InternLM-104B to InternLM-123B in just six months, with strong data processing capabilities. SenseTime possesses a robust SenseCore computational foundation and collaborates with leading research institutions, boasting strong technical expertise.