AI Daily: iOS 27 Supports Free Switching Between ChatGPT; Baidu Wenyin Website Fully Expanded; Google Gemini 3.5 Flash Released

Welcome to the "AI Daily" section! This is your guide to exploring the world of artificial intelligence every day. Every day, we present you with the latest content in the AI field, focusing on developers to help you understand technology trends and innovative AI product applications.

Fresh AI products Click to learn more:https://app.aibase.com/zh

1. Apple iOS 27 Update: Siri Enters the "Dual-Mode" Era, Switching Freely Between Siri AI and ChatGPT

Apple has made a deep architectural upgrade to Siri in the latest iOS 27 system, introducing a new standalone Siri app that supports interaction through text, images, and documents, and allows users to freely switch between Siri AI and ChatGPT, enhancing the intelligent experience.

AiBase Highlights:

Introducing a new standalone Siri application, supporting text, image, and document interaction.

Opening model switching interface, allowing users to freely choose between Siri AI or ChatGPT.

The current version still needs optimization, such as the default mode cannot be permanently set.

2. From Chat Assistant to "Computer Manager": Google Gemini 3.5 Flash Released

The release of Google Gemini 3.5 Flash marks the transformation of AI from a simple Q&A machine to an intelligent agent capable of directly controlling computers. Its powerful computer operation capabilities and security have been significantly improved, providing more efficient solutions for automated office work and tool calls.

AiBase Highlights:

🧠 Gemini 3.5 Flash model optimized computer operation capabilities, able to take over the computer interface directly and complete complex tasks.

🚀 The model outperforms the 3.1 Pro version in coding tasks and autonomous agent tasks, with significantly improved response speed.

🔒 Google enhanced cybersecurity and CBRN risk control to ensure AI agents perform tasks efficiently and securely.

3. Empowered by Wenshin 5.1 Base: Baidu Wenshin Website Fully Upgraded, New Tools Like Office Online Editing Launched

Baidu Wenshin website has been fully upgraded, integrating AI service entry points to create a one-stop super portal. The new site is powered by the Wenshin 5.1 large model, adding tools like Office online editing, and launching AI functions for college entrance exam candidates, covering all scenarios including learning, office work, and daily life.

AiBase Highlights:

🧠 Powered by the Wenshin 5.1 large model, enhancing the technical foundation of AI services.

💼 Added Office online editing function, supporting PPT, Excel, and Word processing.

🎓 Launched AI admission report and AI PPT generation tools for college entrance exam candidates.

4. Sony PS6 Leaks: Equipped with AI Frame Generation Technology

The article details the technological innovations of the Sony PS6 console, including video frame interpolation, super-resolution algorithms, and the use of local AI models, aiming to achieve a high-fluidity 4K 120fps experience and is expected to be released in 2027.

AiBase Highlights:

🎮 PS6 will adopt AMD's improved video frame interpolation technology to enhance visual fluidity.

🧠 Built-in local AI model to reduce latency and optimize ray tracing performance.

💰 Estimated price controlled within $1000, enhancing market competitiveness.

5. Tencent Marvis Ma Weis iOS Version Officially Launched, Achieving Full Platform Coverage and Desktop-Level Remote Control

The launch of the Tencent Marvis Ma Weis iOS version marks significant progress in full platform coverage and cross-end collaboration, offering users a more efficient remote control and office experience.

AiBase Highlights:

📱 Tencent Marvis Ma Weis iOS version officially launched, achieving full platform coverage and desktop-level remote control.

💻 Supports cross-ecosystem connection, breaking down heterogeneous system barriers, and improving multi-device collaboration experience.

🔄 High-frequency iteration R&D, continuously optimizing features, and plans to introduce more deep capabilities in the future.

6. Figma Releases Major Update: Adds Native Code Layers, Animation Support, and Weavy Workflow Generation Functionality

Figma's major update introduces native code layers, animation support, and Weavy workflow generation functionality, significantly improving the collaboration efficiency between designers and engineers and promoting the intelligent transformation of digital product development.

AiBase Highlights:

🎨 Added native code layer to improve collaboration efficiency between designers and engineers.

🎬 Supports animations, transitions, and 3D transformations, simplifying the design process.

🧩 Integrated Weavy workflow generation functionality to enhance the platform's intelligent level.

7. "Chinese Dark Horse" in 3D Generation: Yingshou Technology Secures Billions in Funding, Technical Strength Attracts NVIDIA's Attention

Yingshou Technology has made significant breakthroughs in the 3D generation field. Its next-generation model Rodin Gen-2.5 not only has strong generation capabilities but also achieves high-precision physical-level realistic materials and ultra-high resolution, widely applied in the gaming and film industries.

AiBase Highlights:

🧠 Yingshou Technology released Rodin Gen-2.5, the world's first 3D large model with ten million face-level generation capability.

🔧 The model supports physics-level realistic PBR materials and 12K ultra-high resolution, directly applicable to game engines or film post-production pipelines.

🚀 Yingshou Technology has gained recognition from NVIDIA and other heavyweights due to its solid technology, with core team composed of "00s" generation researchers.

8. Robots Get a "General Brain": RoboScience Launches Visics Large Model, Achieving Cross-Scenario Autonomous Execution

RoboScience released the general embodiment large model Visics, which enhances robots' cross-scenario generalization and autonomous task execution capabilities by introducing "object 3D point cloud trajectory" and a dual-engine architecture. At the same time, its built "simulation + video" dual data flywheel effectively reduces data acquisition costs, providing strong support for the development of embodied intelligence.

AiBase Highlights:

🧠 Visics model introduces "object 3D point cloud trajectory" as a unified intermediate representation standard, enhancing robots' cross-scenario generalization ability.

🔄 Uses a dual-engine architecture, where the embodied world model and general operation model work together to achieve understanding of the physical world and conversion of control instructions.

📊 Builds a "simulation + video" dual data flywheel, significantly reducing data acquisition costs, and promoting the development of embodied intelligence.