Tencent Cloud today unveiled a suite of productivity agent tools at the Tencent Cloud AI Industry Application Conference in Beijing. The new suite offers differentiated and personalized solutions for AI agent development and deployment, addressing the growing demand from individuals and enterprises to adopt AI across more than 20 vertical scenarios. 

For individual users, Tencent Cloud has upgraded a range of out-of-the-box AI tools, including QClaw, WorkBuddy, Yuanbao, ima, and Tencent Docs, making productivity more accessible than ever. For enterprise clients, Tencent Cloud unveiled the all-new WorkBuddy Enterprise AI Workspace, and upgraded its core products such as ClawPro, the Tencent Cloud Agent Development Platform (ADP), and Qidian Marketing Cloud, and more building a comprehensive, full-lifecycle service ecosystem.

Dowson Tong and Shunyu Yao Share the Stage for a Dialogue on AI

During the conference, Dowson Tong, Senior Executive Vice President of Tencent was joined by Tencent’s Chief Scientist for AI, Shunyu Yao, in a keynote conversation on the next phase of AI development and Tencent’s AI strategy.

Tong said that Tencent’s AI strategy has always been grounded in practicality, usability and scalability. "The key is to build AI around real-world scenarios. These scenarios capture real user needs and provide the data needed for models to keep improving. With its broad product portfolio, interactive data and ecosystem connections, Tencent’s scenarios provide high-quality context, enabling AI to invoke tools, connect with systems and complete tasks from end to end."

Yao said AI is entering a new phase, with the focus shifting from finding the right methods to identifying the right problems to solve. At the center of this shift, it is a balanced triangle framework of “foundation–product–frontier”. He added that what sets large language models apart is their ability to generalize across tasks and scenarios, enabling data and insights from different use cases to reinforce one another. As AI agents emerge, the interaction between environments, evaluation, reasoning and action is becoming increasingly important even more so than model capability alone. This underpins Tencent’s renewed approach to Hunyuan, which emphasizes on co-design between models and products, as the model continues to improve in both scale and performance.

A Conversation on Co-Designing Models and Products

Setting a New Standard for AI-Powered Efficiency

As AI enters its next phase, agents are emerging as an important engine for reshaping productivity. Tencent Cloud’s Productivity Agent Suite brings together a broad portfolio of AI agent tools and solutions, spanning ready-to-use tools, enterprise-grade platforms and industry solutions. Designed around user needs, the suite helps individuals and enterprises apply AI more easily across work, development, collaboration and industry-specific scenarios.

For individual users, the out-of-the-box AI tools across its product ecosystem, include AI assistants for knowledge management, document processing, content creation and everyday productivity. These tools are designed to make AI easier to access and apply in daily work and learning scenarios.

For professionals and workplace users, Tencent has expanded its “Buddy” family of AI tools to address high-frequency productivity needs across software development, document processing, creative design and collaboration. WorkBuddy, Tencent’s personal AI agent workspace, has become one of China’s most widely used productivity agent tools by daily active users. CodeBuddy, Tencent’s coding agent, supports plug-in, IDE and CLI formats, covering key software development workflows such as coding, code review, testing and operations. Tencent Miora and Ardot are designed for creative generation and interactive design, helping professionals improve efficiency in specialized workflows.

For enterprise clients, Tencent Cloud unveiled WorkBuddy Enterprise and an office AI Agent Suite that support human-AI collaboration. The office AI Agent Suite also connects Tencent Docs and Tencent LearnShare into the WorkBuddy workspace through One ID, creating an AI-native environment that enables the full workflow from content creation and knowledge accumulation to capability sharing.

With Tencent Docs integrated into WorkBuddy, users and AI agents can co-create and collaborate within the same document. Tencent Meeting also uses AI to capture and structure meeting context, enabling different AI agents to understand and retrieve meeting content through interfaces such as Model Context Protocol (MCP), Skills and Command-Line Interface (CLI).

Tencent Cloud also upgraded its agent management and development platforms, including ClawPro and Tencent Cloud Agent Development Platform (ADP). The latest ADP 4.0 is positioned as an enterprise-grade AgentOps platform, covering the full lifecycle of AI agent development, connection, distribution and governance. 

Tencent’s Productivity Agent Suite is built on the company’s internal AI transformation experience. CodeBuddy has helped reshape Tencent’s software development workflow, supporting more than 95% of Tencent engineers and reducing overall coding time by 40%. WorkBuddy has enabled closer collaboration between employees and AI agents, helping small teams build and iterate products more efficiently.

Upgrading Cloud Platforms and Capabilities for AI Agent Deployment

Tencent Cloud has upgraded its cloud product portfolio to support the large-scale deployment of AI agents in real business scenarios. The full-stack capabilities span models, inference services, agent infrastructure and ecosystem connectivity, helping lower the barriers to agent development and accelerate adoption in production environments.

Its latest Hy3 Preview model has been optimized for agent task execution and cost performance, and is already being integrated across products such as CodeBuddy, WorkBuddy and Yuanbao. In WorkBuddy, Hy3 Preview improved first-response speed by 54%, reduced average task completion time by 47%.

Tencent Cloud’s TokenHub platform is designed to make large model inference more efficient and easier to manage at scale. It integrates Tencent Hunyuan and mainstream third-party models, with support for Day-0 model access, unified protocol and key management, and dynamic routing based on cost, performance and intelligence. The platform is now available globally.

Tencent Cloud has also upgraded its Agent Runtime, providing native infrastructure for agent operation, memory, storage and secure access. With elastic scheduling and on-demand scaling, the runtime can release up to 70% of idle computing resources, while its memory service reduces token consumption by 60% and improves task success rates by 30%.

Today, Tencent’s Productivity Agent Suite has already been applied across more than 20 industries, including healthcare, consumer electronics, finance, gaming, retail and education. By bringing AI agents into real-world business scenarios, Tencent Cloud is helping leading enterprises transform how teams work and build stronger human-AI collaboration at scale.