Search AI Products and News

Tao Tian Group announced yesterday at its 'Hard-Core Youth Tech Festival 4.0' event that its independently developed hundred-billion-parameter recommendation large model, RecGPT, has officially gone live. This innovative achievement will comprehensively upgrade the 'You May Also Like' information flow on the mobile Taobao homepage, providing users with a more accurate and personalized recommendation experience through AIGR (Generative Recommendation) technology. The launch of RecGPT marks an important step forward for Taobao in the field of e-commerce recommendations. Test data shows that the recommendation feed equipped with the RecGPT large model performs excellently, with real user click-through rates

On June 30th, the world's first trillion-parameter large model for the power generation industry, Qinyuan, developed independently by the State Energy Group, was officially released. This innovative large model, with its unique full-stack independent and controllable characteristics, marks an important step forward for the power generation industry into a new era of intelligent decision-making. The Qinyuan large model aims to integrate the diverse scenarios, high complexity, and strong professionalism of the power generation industry, fully utilizing the State Energy Group's globally largest installed capacity and massive data resources. The launch of this model not only creates a new engine for multi-energy coordination and dynamic optimization, but also establishes an electricity

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.

China National Petroleum Corporation Limited officially released its latest Kunlun large model with 300 billion parameters in Beijing. This significant achievement marks another important breakthrough for China in the field of large model technology, demonstrating the rapid progress and innovative capabilities of state-owned enterprises in the field of artificial intelligence. The new Kunlun large model has made remarkable improvements compared to the previous 700-billion-parameter version released in November 2024. The parameters of the language large model increased from 700 billion to 300 billion, the parameters of the vision large model increased from 3 billion to 44 billion, while those of the multimodal large model were...

Tencent has released the technical report on Qwen TurboS, revealing the core innovations and powerful capabilities of its flagship large language model TurboS. According to the latest rankings from the global authoritative large model evaluation platform Chatbot Arena, Qwen TurboS ranks seventh among 239 participating models, making it the top model domestically after Deepseek, and internationally trailing only Google, OpenAI, and xAI.

Developed by TNG Technology Consulting, the DeepSeek R1T Chimera model has officially launched on the OpenRouter platform, providing global developers with efficient and powerful inference capabilities. This new open-source model combines the excellent inference capabilities of DeepSeek R1 with the high performance of V3-0324, marking another significant breakthrough in the balance of performance and efficiency in open-source AI technology. The following is compiled by AIbase.

The Ostris team has released Flex.2-preview, an 800 million parameter text-to-image diffusion model designed for integration into ComfyUI workflows. According to AIbase, this model excels at generating images with strong control over lines, poses, and depth. It supports general control and inpainting functionality, continuing the fine-tuning evolutionary path from Flux.1Schnell to OpenFlux.1 and Flex.1-alpha. Flex.2-preview is available on Hu...

Xiaopeng Motors chairman He Xiaopeng recently announced that the company's fully self-developed Turing AI chip will be mass-produced and launched in the second quarter of this year. This progress comes as the automotive industry accelerates the application of end-to-end intelligent driving technology and the scale of AI large models continues to expand. Xiaopeng Motors is building its strongest AI brain by simultaneously developing a world base model with 35 times the parameters of mainstream VLA models, and a self-developed chip with computing power equivalent to three Nvidia Orin Xs, which is about to be mass-produced.

XPeng Motors held an AI technology sharing session in Hong Kong yesterday, unveiling its newly developed 720-billion-parameter large-scale autonomous driving model – the "XPeng World" base model. This model features a multi-modal architecture, integrating visual understanding, chain reasoning, and action generation capabilities. It aims to be deployed to vehicle-end devices via cloud distillation technology and simultaneously empower ecological products such as AI robots and flying cars.

Recently, the AI code generation field has seen a surge in open-source releases, with several heavyweight models making their debut. Among them, the Cogito v1 Preview series from Deep Cogito stands out. According to AIbase, this new family of open-source models includes various sizes: 3B, 8B, 14B, 32B, and 70B parameters. Not only does it outperform competitors in its class, but its 70B version even surpasses Meta's recently released Llama 4 109B MoE model, sparking considerable industry discussion.

JD Retail's technology team announced the successful launch of TimeHF, its first self-developed billion-parameter time series model for sales forecasting. This model leverages Reinforcement Learning from Human Feedback (RLHF), applying it for the first time to sales forecasting. The accuracy of predictions has been significantly improved by over 10%, substantially reducing uncertainty in demand forecasting. This achievement has demonstrated excellent performance in JD's internal automated replenishment scenarios for 20,000 products.