Latest news with #Skywork
Malaysian Reserve
4 days ago
- Business
- Malaysian Reserve
Skywork Deep Research Agent Major Upgrade: Delivering Enhanced Multimodality, Superior Output Quality, and Optimized Efficiency
SINGAPORE, Aug. 14, 2025 /PRNewswire/ — The SkyWork AI Technology Release Week officially kicked off on August 11. From August 11 to August 15, SkyWork will release one new model each day for five consecutive days, covering cutting-edge models for core multimodal AI scenarios. As of now, we have already launched the SkyReels-A3, Matrix-Game 2.0, Matrix-3D, and Skywork UniPic 2.0 models. On August 14, Skywork Deep Research Agent v2 was officially launched as the core engine of Skywork Super Agents. Since its initial launch on May 22, Skywork Deep Research Agent has significantly reshaped the role of large language models in the AI Office space. Through the platform, it has produced a vast number of high-quality documents, PowerPoint presentations, spreadsheets, and other deliverables with exceptionally high information density for users. The newly upgraded Skywork Deep Research Agent v2 introduces the following enhancements to the user experience. Users worldwide are welcome to register and use website: 1 'Multi-Modal Deep Research' Agent – The First to Integrate Multi-Modal Retrieval, Understanding, and Generation Existing Deep Research Agents in the industry rely exclusively on searching and scraping textual data from web pages, restricting their analysis to plain text. However, more than half of the internet's critical information exists in mixed text-and-image formats, such as financial report graphs, experimental diagrams in research papers, social media comparison visuals, and proposal flowcharts. Overlooking such multi-modal data deprives the Agent of key decision-making insights, significantly compromising output quality. To address this challenge, the Skywork team has launched the industry's first Multi-Modal Deep Research Agent by seamlessly combining multi-modal retrieval, understanding, and cross-modal generation into deep research workflows. This feature is now live on ( and available to users worldwide. To enhance multi-modal information retrieval capabilities, the Skywork team has pioneered four technological breakthroughs, including MM-Crawler (multi-modal crawler) technology, long-context multi-modal information aggregation, asynchronous parallel processing with multi-agent understanding architecture, and multi-modal output generation. Through these technological innovations, the multi-modal Skywork Deep Research Agent v2 has finally accomplished what seems simple yet was long neglected—simultaneous text reading and image comprehension. Therefore, it enables researchers and users to generate comprehensive, logically structured, and visually refined in-depth reports in a single step. 2 'Multi-Modal Deep Browser Agent' – Redefining Social Media Analytics and Data Intelligence To deliver capabilities unmatched by conventional browsers, including ultra-low latency, guaranteed high response rates, optimized task completion, and adaptive decision-making, the team has implemented critical proprietary advancements in the Skywork Browser Agent, covering enhanced DOM + visual reasoning architecture, native integration with major platforms, Parallel Search technology, Multi-Action planning mechanism, intelligent filtering, seamless human-AI collaboration, privacy protection and security compliance. The Skywork Browser Agent now achieves human-like browsing and interaction capabilities, fundamentally transforming traditional approaches to data collection and analysis. The agent demonstrates remarkable precision and efficiency in executing intelligent search operations, performing multimodal information analysis, and deriving actionable insights from community content. By effectively resolving limitations inherent in conventional browser agents, it showcases the significant potential of Skywork Super Agents in handling both long-horizon tasks and vision-language actions (VLA). The Skywork Browser Agent has entered its alpha and invite-only testing phase, with full public release expected soon for all users. Skywork Browser Agent's core capabilities: Advanced multimodal comprehension: Going beyond text-only analysis, it achieves deep semantic understanding of social media content—including platforms like Xiaohongshu, Twitter, and Instagram—by extracting insights from images/videos and analyzing comment sentiment, enabling holistic data intelligence. Automated data analysis & reporting: The agent performs efficient community content analysis and transforms raw research data into intuitive, visually digestible reports. One-click website generation: The agent automatically curates key visuals, analyzes their content, and deploys them as ready-to-use standalone websites with one click, thereby streamlining result presentation and team collaboration. Seamless workflow integration: It is designed for interoperability with information retrieval agents and document tools (e.g., PPT/Doc assistants). When drafting reports, it intelligently retrieves and recommends relevant visual assets, thus dramatically boosting productivity. 3 Achieving SOTA Across Various Benchmarks with Enhanced Deep Information Retrieval & Complex Task Execution Capabilities To enhance the foundational model's performance in complex task execution, including advanced information retrieval, synthesis, and summarization, Skywork Deep Research Agent v2 integrates multiple breakthrough mechanisms: high-quality synthetic data generation and curated training, end-to-end reinforcement learning, highly efficient parallel inference, and multi-agent self-evolution frameworks. Benchmark evaluations confirm its superior performance, setting new state-of-the-art (SOTA) results industry-wide. On the authoritative search evaluation benchmark BrowseComp, Skywork Deep Research demonstrates exceptional performance. In standard mode, it already surpasses most competing solutions with an accuracy rate of 27.8%. When activating its proprietary Parallel Thinking mode, the accuracy jumps significantly to 38.7%, setting a new industry SOTA record. Notably, in Parallel Thinking mode, Skywork Deep Research's accuracy exhibits continuous improvement with extended processing time, which demonstrates the exceptional scalability and untapped potential of our proprietary system architecture. The API preview feature is now available. To request access, please visit Skywork's official GitHub repository and submit your application: In addition, Skywork Deep Research Agent has achieved SOTA performance on the GAIA Test benchmark, which validates its advanced capabilities in complex task execution Skywork Deep Research Agent v2 will soon launch comprehensively across all deep research applications on
Malaysian Reserve
6 days ago
- Entertainment
- Malaysian Reserve
Matrix-Game 2.0 Launches as a Powerful Open-Source Alternative to Genie 3
SINGAPORE, Aug. 12, 2025 /PRNewswire/ — The SkyWork AI Technology Release Week officially kicked off on August 11. From August 11 to August 15, a new model will be unveiled each day, covering cutting-edge models for core multimodal AI scenarios. A week ago, DeepMind released a major update to its interactive world model—Genie 3—enabling real-time, long-sequence generation. This advancement has drawn significant attention to world models. However, Genie 3 was not open-sourced, leaving the community to speculate about its implementation. On August 12, Skywork unveiled an upgraded version of the self-developed Matrix series' interactive world model—Matrix-Game 2.0. It also delivers interactive, real-time, long-sequence generation in general scenarios. To drive progress in interactive world modeling, Matrix-Game 2.0 has been fully open-sourced, marking the industry's first open-source solution for real-time, long-sequence, interactive generation in general scenarios. Matrix-Game 2.0 open source addresses: Technical report: Project homepage: HuggingFace: GitHub: Matrix-Game 2.0 achieves a breakthrough in real-time generation and long-sequence handling. Compared to its predecessor, the 2.0 version prioritizes low-latency, high-frame-rate performance for extended interactions, enabling stable 25 FPS continuous video generation across complex scenes. Its generation length scales to minute-long sequences, drastically improving temporal coherence and real-world usability. While delivering a significant boost in inference speed, Matrix-Game 2.0 maintains precise comprehension of physical laws and scene semantics. It enables users to freely explore, manipulate, and construct virtual environments in real time through simple instructions—yielding well-structured, detail-rich, and logically coherent virtual spaces. With these capabilities, Matrix-Game 2.0 not only breaks down the barriers between content generation and interaction but also unlocks new possibilities for cutting-edge applications such as virtual humans, game engines, and embodied AI. It provides a robust technical foundation for building a universal virtual world. Currently, Matrix-Game 2.0 boasts three core advantages: High-frame-rate, real-time long-sequence generation: The model supports fluid movement (forward/backward, left/right) and camera/view rotation. Users can intuitively control characters in the scene via simple commands. The system generates seamless footage in real time at 25 FPS, enabling minute-long interactive sequences in a single session. Character movements are lifelike, smooth, and precisely responsive. Cross-scenario generalization capability: The model demonstrates exceptional cross-domain adaptability. It is not only suitable for specific task scenarios but also supports simulations of diverse styles and environments—including urban, wilderness, and other spatial types, as well as realistic, oil-painting, and various visual styles. Enhanced physical consistency: The model demonstrates a deeper understanding of physical rules. Characters generated by the model exhibit physically plausible movements when navigating complex terrains such as steps and obstacles, which improves immersion and controllability. The open-source release of Matrix-Game for interactive video generation underscores Skywork's strategic foresight in AI development. This initiative will accelerate development across Skywork's multi-model AI ecosystem. Moving forward, Skywork remains committed to pioneering and open-sourcing advanced AI solutions. By collaborating with global developers and users, we aim to build next-generation platforms that accelerate the global advancement of AGI.
Yahoo
22-05-2025
- Business
- Yahoo
Skywork Launches Skywork Super Agents Globally: The "AI-Powered Office Suite" Built on Deep Research
SAN FRANCISCO, May 22, 2025 /CNW/ -- On May 22, Skywork ( unveiled Skywork Super Agents to the global market. This cutting-edge product leverages AI agent architecture and deep research technology to provide an all-in-one solution for generating documents, slides, Excel sheets, webpages, podcasts, and multimodal audiovisual content. With its industry-leading deep research capabilities, Skywork Super Agents ranked #1 globally on the GAIA benchmark, surpassing competitors such as OpenAI Deep Research and Manus.[1] The launch of Skywork Super Agents revolutionizes traditional office software - heralding the dawn of the "AI Office Agent" era. Available immediately, users worldwide can now register and access the platform without requiring any invitation code. • Global official website: 01 5 Modes, One Click—Get 8 Hours of Work Done in Just 8 Minutes As a working professional, you have surely felt this: 60% of your time is eaten up by drafting complex materials. You are stuck compiling endless research reports, summaries, data sheets, and presentations—cycling through "final versions," "truly final versions," and "absolutely final versions." You have burned the midnight oil too often, with no time left to create real value. Those days are over. Introducing Skywork Super Agents—the pioneer in "AI Office Agents." With 5 expert-level agents and 1 general agent, it supercharges your content creation, delivering instantly high-quality, ready-to-use materials you can trust. Unlike other AI agents which offer broad functionality but lack specialized depth, Skywork Super Agents have built a vertically specialized system consisting of "5 expert agents" and "1 general agent." 5 Expert Agents. Each specializes in generating professional documents (Docs), PowerPoint slides, Excel sheets, podcasts, and webpage content. Among these, documents, slides, and sheets—often called the "Office Trio"—are the most essential tools for professionals, forming the core of platforms like Microsoft Office and Google Workspace. Skywork's three Office-focused agents integrate OpenAI Deep Research-level capabilities, delivering expert-grade, consultancy-level, and research-ready outputs. Meanwhile, the webpage and podcast agents cater to the dynamic needs of the digital age, offering engaging and meaningful content formats. All five expert agents are designed for real-world office and academic scenarios, providing users with tailored, high-quality content. 1 General Agent. Powered by dozens of MCPs (Multimodal Creative Processors), it excels at handling multimodal creative tasks, including images & posters, music & music videos (MV), promotional videos, audiobooks, illustrated books, and other multimedia content. Therefore, Skywork Super Agents empower users across industries to create trustworthy, editable, and ready-to-use content, transforming AI's role from a mere assistant into a true productivity partner! 02 Deep Research + Office Trio: Skywork's Two Core Competencies The centerpiece of this launch is Skywork Super Agents' three flagship AI agents: Documents, Slides, and Sheets. First, Documents. Modern users confront increasingly sophisticated and varied writing needs, spanning industry research, competitive analysis, product roadmaps, academic papers, business plans, marketing collateral, and creative copywriting. These cross-disciplinary demands - bridging business strategy, academic rigor, and marketing impact - require content that combines expert-level precision, innovative thinking, and immediate applicability. To meet these needs, Skywork has integrated advanced deep research capabilities into its "Documents" agent. The company has independently developed a deep research model that enables intelligent information retrieval, leveraging the model's advanced reasoning and deep thinking abilities to enhance both the scope and depth of searches while boosting efficiency. Through reinforcement learning, the model's search capabilities are further generalized, ensuring high-quality source information for user-generated content. Skywork delivers deep research performance on par with OpenAI's, at just 40% of the cost. Skywork's deep research agent framework scored 82.42 on the GAIA benchmark, outperforming OpenAI Deep Research and Manus to claim the top spot (data as of May 10, 2025).[1] On SimpleQA, Skywork achieved a score of 94.5, surpassing the current state-of-the-art (SOTA).[1] Compared to OpenAI's Deep Research, the research reports generated by Skywork's "Documents" agent feature a more diverse range of data visualizations, including bar charts, histograms, line graphs, pie charts, radar plots, and data tables. These elements are presented in a visually engaging and dynamic format within Skywork-generated reports. For example, we task Skywork Super Agents with creating an industry research report: "I'm planning to launch a tech company specializing in wearable health monitors for pets. Could you generate a comprehensive industry report on the pet care market, with as many visual charts and data insights as possible?" ("I am currently interested in wearable devices for pets. Please generate a market growth and industry research report on the pet wearable device market. The report should include key trends, growth drivers, and challenges in the industry, with supporting data wherever possible. Please also include visualizations including line graphs, bar charts, timeling charts and radar charts to illustrate the key findings and trends. The chart's color scheme should primarily feature a blue-green palette. The design should have a unified tone and a modern feel.") Not limited to "Documents," Skywork's "Slides" agent also harnesses deep research capabilities. Following comprehensive searches and analysis, it produces both accurate and visually impressive content. Every presentation fact and data point is source-verifiable, while the slides themselves boast engaging designs, refined aesthetics, and dynamic effects. The slides support online editing and export to both PPTX and PDF formats. For instance, when we task Skywork Super Agents with creating a tourism marketing presentation for Iceland highlighting its natural attractions - "Please generate a slide deck on Iceland's tourism marketing, focusing on promoting Iceland's natural attractions", it immediately generates a professional slide deck. With just a click on the top-right "Edit" button, we can freely customize all content. Skywork's "Sheets" agent also supports deep research. It can perform descriptive or inferential statistics based on user-uploaded data tables and generate statistical charts. It is also adept at creating various "template" or "summary" tables. The generated tables can not only be viewed online but also exported as offline documents in XLS format. For example, we upload - "Please analyze the correlation between product sales and discounts for the three products based on the past 8 weeks of sales data provided in the attachment. It would be great if you could generate visual charts for better understanding." Skywork Super Agents clearly outlines the tasks and generates the corresponding charts as requested. To give back to the developer community, Skywork has made its deep research agent framework open-source, now available for download on Hugging Face. Additionally, Skywork has integrated its document, slide, and sheet generation capabilities into the MCP for developers. 03 Webpages, Podcasts & Multimedia: Skywork Super Agents' Multimodal Generation Makes It an All-in-One Powerhouse Skywork Super Agents is not just an "AI-powered Office," but also a versatile expert in multimodal content creation. It transcends single-format limitations, extending generation capabilities to webpages, podcasts, and multimedia, and offering a one-stop content creation ecosystem. The "Webpages" agent can quickly create well-structured, highly interactive professional websites based on user needs. From e-commerce pages to personal blogs, from informational displays to functional websites—no complex coding required. It brings creative visions to life effortlessly. Skywork's "Podcasts" agent can generate logically compelling and engaging scripts from just a simple sentence. Building on these scripts, the agent leverages voice synthesis technology to produce high-quality audio with diverse vocal tones. Currently, Skywork supports English podcast generation, with Chinese and other languages coming soon. Additionally, Skywork's "General" agent integrates multiple MCPs, including image generation, video generation, music composition, and voice synthesis. The agent can instantly convert text into polished videos with seamless editing, complete with automatically matched background music and effects. From promotional videos and educational content to creative short films, audiobooks, and illustrated books, it handles all content types effortlessly. 04 Smart, Reliable, and Knowledge-Enriched: Ingenuity in the Design of Skywork Super Agents Skywork Super Agents not only delivers top-tier generative AI capabilities but also pushes the boundaries of user experience. The product and design teams have infused the platform with multiple innovations, making these agents smarter, more reliable, and knowledge-enriched, and transforming them into truly useful, trustworthy, and intuitive AI agents. Smart | Automated Request Clarification Industry pain point: Jumping into execution without fully understanding requirements The AI Agent sector has been grappling with "blind execution" during its development. Many AI Agent products demonstrate inadequate semantic parsing of prompts, implicit requirements, and application scenarios when processing user instructions. Relying solely on surface-level keyword extraction, they hastily initiate tasks, failing to discern users' underlying intent and leading to contextually irrelevant outputs. Sometimes they even misinterpret core instructions—ultimately delivering outputs that completely fall short of users' expectations. Smarter tasks begin with smarter inputs. Skywork Super Agents doesn't just respond—it truly "listens." Breaking away from vague prompt execution, it pioneers the "Clarification Card" feature. Before initiating tasks, it inquires about users' objectives, context, and constraints by having them complete fill-in-the-blank and multiple-choice questions. It then generates a "To-do list" for users to double-confirm, ensuring precision from the very start. Leaving no critical checkpoint unchecked, Skywork Super Agents precisely understands users' real needs before generating content. This guarantees stronger control over key actions, so that the output could better align with expectations. Reliable | Traceable Sources Industry pain point: Data based on AI hallucinations—unreliable results AI hallucination has become a critical pain point in the industry's development. Some AI systems, lacking rigorous mechanisms to verify the authenticity of knowledge, frequently fabricate data out of thin air. They quickly produce seemingly complete results, though these outputs are often riddled with misinformation and false claims—severely undermining their credibility and reliability. Skywork Super Agents tackles this problem with its "source tracing" feature. Every piece of content generated its Documents, Slides, and Sheets agents comes with clearly traceable references, allowing users to verify information directly from its original context. This level of transparency is critical for professionals who need to defend their viewpoints, researchers pursuing precision, and students who must cite reliable sources. With Skywork Super Agents, users can create and deliver content with confidence—no second-guessing content validity. The output isn't just usable; it is ready to trust and deploy from the start! Knowledge-Enriched | Building a Personal Knowledge Base Industry pain point: No accumulation of materials and results Similar to NotebookLM, Skywork Super Agents allows users to build their "personal knowledge bases." Users can upload files in various formats—PDF, DOC, PPT, XLS, and more—as well as audio recordings or URL links. Each knowledge base supports up to 50 documents. Additionally, users can create multiple personal knowledge bases, the content of which can be used to generate documents, slides, sheets, web pages, and podcasts. This feature enables users to quickly generate content from existing materials and effortlessly reuse knowledge they've accumulated—streamlining research, boosting efficiency, and creating a smarter, continuously evolving workspace. The launch of Skywork Super Agents marks a pivotal leap in AI technology—from single-function tool development to productivity empowerment across all scenarios. Beyond delivering a "what-you-think-is-what-you-get" intelligent experience, it establishes new technical benchmarks that drive industry-wide innovation across model optimization, tool integration, and scenario adaptation. This progress incentivizes enterprises to increase AI-related investments, subsequently activating the upstream/downstream of the industry chain and accelerating the development of a thriving, symbiotic AI ecosystem. Let's try it now! Skywork Super Agents: • Global official website: • Deep research agent open-source address: • MCP address: Note: [1] ranked #1 on May 10, 2025, refer to the article on Kunlun Wanwei Group's official account View original content to download multimedia: SOURCE Skywork AI pte ltd View original content to download multimedia:
Cision Canada
22-05-2025
- Business
- Cision Canada
Skywork Launches Skywork Super Agents Globally: The "AI-Powered Office Suite" Built on Deep Research
SAN FRANCISCO, May 22, 2025 /CNW/ -- On May 22, Skywork ( unveiled Skywork Super Agents to the global market. This cutting-edge product leverages AI agent architecture and deep research technology to provide an all-in-one solution for generating documents, slides, Excel sheets, webpages, podcasts, and multimodal audiovisual content. With its industry-leading deep research capabilities, Skywork Super Agents ranked #1 globally on the GAIA benchmark, surpassing competitors such as OpenAI Deep Research and Manus. [1]
