Search AI Products and News

Midjourney adds 'For You' feature, offering personalized AI-generated content recommendations based on user interaction history (likes, moodboards). Located top-left, it enables one-click access to style-matched images/videos, adjustable via stylize/chaos parameters for precise genre alignment.....

The Qwen3-30B-A3B model has undergone a major update, launching the new version Qwen3-30B-A3B-Thinking-2507. This new version has achieved significant improvements in reasoning capabilities, general abilities, and context length, marking that the model is not only lighter but also has stronger thinking capabilities. Compared to the Qwen3-30-A3B model open-sourced on April 29, the new model has made progress in multiple key areas. In terms of mathematical abilities, Qwen3-30B-A3B-Thin

Google has launched a new AI search mode in the UK, powered by the Gemini 2.5 model, which supports complex questions and can handle multi-part queries. The new feature uses query expansion technology to break down questions into sub-topics for parallel search, providing more accurate answers. It supports multiple interaction methods, including text, voice, and images, allowing users to upload images or ask questions via voice. The AI mode evaluates the credibility of the answers, and if the confidence is low, it switches to traditional search results. This feature aims to improve the depth and efficiency of searches while promoting diversified website traffic. It is still under optimization, and Google encourages users to provide feedback on their experience.

Google's NotebookLM introduces video summaries, converting complex info into narrated slides. It auto-generates videos from user-uploaded images/charts. Demoed at I/O 2025, the feature will roll out gradually. Currently English-only, multilingual support planned.....

Microsoft Edge launches Copilot mode, using AI to redefine browsing. It automates complex searches (e.g. housing, price comparison), integrates multi-source data for precise results. Features natural voice interaction, smart recommendations from open tabs, and a unified interface combining search, chat & navigation. Ensures strict privacy with fully controllable data usage. Currently free, with on/off toggle.....

Perplexity company launches an AI browser called Comet, which transforms YouTube videos into interactive learning tools. Users can pause videos at any time to ask questions, and the AI assistant provides instant answers and supplements related knowledge, enabling a shift from passive watching to active learning. This tool is especially suitable for complex subjects, as it optimizes answers based on user habits and offers personalized learning paths. Its deep integration with the browser environment makes responses more contextually relevant. Although still in the testing phase, it has already shown the potential to revolutionize the way education is delivered.

Tencent launches the world's first open-source 3D world generation model, Hunyuan 3D 1.0, supporting 360° high-precision scene generation and interactive exploration. This model significantly lowers the barrier to 3D creation through AI technology, enabling immersive scenes to be generated quickly from a single image or text. It offers 20 free experiences per day. Its open-source nature empowers developers worldwide and is applicable to multiple fields such as gaming, VR, and film. The project has over 12,000 stars on GitHub. This technology advances 3D generation from object-level to world-level, redefining content production processes. It has the potential to drive innovations in gaming and digital worlds in the future.

Google introduces 'Web Guide' using Gemini AI to reorganize search results with smart categorization and AI summaries. Currently in testing, it enhances search efficiency with contextual answers.....

OpenAI launched ChatGPT's agent feature for paid users on July 24, 2025, integrating Operator and Deep Research tools to enable AI to autonomously handle multi-step tasks like travel planning and financial reports. It excels in benchmarks with 27.4% math accuracy. Plus/Pro/Team users can access it now, with enterprise rollout coming.....

LiblibAI's Lovart AI launches globally as the first commercial AI design Agent, serving 800K users across 70 countries. It integrates top models via a 'Creative Reasoning Engine' to generate 40 professional designs from text, covering branding and marketing. Features include ChatCanvas for natural language adjustments and learning user preferences. A China-optimized 'Stellar Agent' enhances Chinese semantic processing. The tool lowers design barr....

XREAL launches AR smart glasses XREAL One Pro, using its self-developed X Prism optical engine technology, achieving a 57° ultra-wide field of view, increasing the visible area by 63%. Equipped with X1 spatial computing chip, 3ms ultra-low latency, compatible with multiple devices. Over 10,000 pre-orders have been obtained overseas, with a domestic price of 4299 yuan, shipping in mid-July.

Zhipu Z.ai launches Zread.ai, a Chinese open source project reading tool that supports pasting GitHub links to automatically generate project structure and usage guides, significantly lowering the barrier for developers to understand open source projects. The tool has indexed numerous popular projects and offers an application process for less popular projects to be indexed. Its feature Buzz aggregates community updates, including commits, issues, and news, helping developers stay fully informed about project progress. This tool fills the gap in Chinese open source project reading tools and is expected to become an essential tool for developers.

Google upgrades its AI note-taking app NotebookLM with a new public note library. The library integrates authoritative content such as The Economist and expert notes, allowing users to ask questions and receive AI summaries. New features include audio overviews and mind maps, with initial resources covering practical topics such as longevity guides and travel tips. The platform has currently accumulated over 140,000 user notes, leveraging AI technology to improve knowledge management efficiency and strengthen the information sharing ecosystem.

AI voice input platform Willow Voice raised $4.2M in angel funding from Y Combinator. Its tool converts speech to text in real-time for macOS, with features like terminology recognition and privacy protection. Funds will boost R&D and cross-platform expansion.....

The neighborhood social application Nextdoor has launched a major update with three new features: local news aggregation (collaborating with 3,500 local media outlets), real-time safety alerts (weather, traffic, and disaster warnings), and AI-powered Faves recommendations (personalized local service suggestions generated based on 15 years of neighborhood data). The platform aims to address previous issues with incorrect information by improving content quality and timeliness to revitalize user engagement. CEO Tolia emphasized that its core advantage lies in digital neighborhood reputation, which is a unique value that other platforms cannot replicate.

Blok is a startup that focuses on AI testing tools. Its innovative technology can simulate user roles for application testing, helping developers predict user behavior in advance. The founding team consists of experienced entrepreneurs and has raised 7.5 million USD in funding. Compared to traditional testing tools, Blok is more forward-thinking and can provide improvement suggestions before coding. The product is currently in closed beta, primarily serving industries such as finance and healthcare, which require high testing accuracy. It is expected to generate millions of dollars in revenue this year.

Microsoft open-sources the Phi-4-mini-flash-reasoning model, specifically designed for edge devices, with inference efficiency improved by 10 times. It uses an innovative SambaY architecture to achieve efficient memory sharing, showing outstanding performance in long text generation and mathematical reasoning. Benchmark tests show its excellent long context understanding ability, with a Phonebook task accuracy rate of 78.13%. This model is suitable for educational and research fields and can run on a single GPU.

Tencent open-sourced 'Hunyuan-A13B', an 80B-parameter MoE model with dynamic inference (130B active params). Supports 256K context, trained on 20T tokens. Achieves 87.3% on AIME2024 math, but comparisons show version disparities.....