Workbuddy's "Openness and Win-Win" Approach: Harness Becomes Tencent's "Crown Jewel"

Workbuddy's "Openness and Win-Win" Approach: Harness Becomes Tencent's "Crown Jewel"

Tencent's AI workbench, WorkBuddy, has officially opened its ecosystem platform, showcasing its platform ambitions in the agent sector with a lineup of over 100 partners. Jefferies' latest research report characterizes Harness as Tencent's "crown jewel," believing its flywheel effect is emerging and its monetization potential is vast. It maintains a buy rating on Tencent Holdings with a target price of HK$750, implying approximately 70% upside from the current share price.

The WorkBuddy open platform was officially launched recently, initially introducing over 100 ecosystem partners, covering three levels: smart hardware, industry applications, and developers. It has become the first open platform in China's AI Agent sector to simultaneously connect these three ecosystem layers. Liu Yi, Vice President of Tencent Cloud and head of Tencent CodeBuddy and WorkBuddy, stated on September 2nd, "We hope to build WorkBuddy into an operating system for the Agent era—not just a more powerful tool, but a platform that can support all tools."

In a subsequent research report, Jefferies noted that Tencent, at a recent institutional investor conference, explicitly positioned Harness as a core strategic asset and emphasized its continued commitment to consolidating its leading position. Analysts Thomas Chong and Zoey Zong believe that WorkBuddy possesses significant advantages across multiple dimensions, including addressable market size, skill ecosystem density, and model collaboration efficiency, and its current valuation does not fully reflect its long-term potential.

Open Platform Strategy: Fully Connecting the Three-Tier Ecosystem

Since its launch in March this year, WorkBuddy has completed more than 50 iterations and upgrades. This ecosystem open event marks the entry of its platform strategy into a substantial implementation stage.

At the industry application level, WorkBuddy has established partnerships with over 20 professional industries, covering fields such as finance, SaaS, law, healthcare, education, public welfare, and design. It also integrates with enterprise systems like CRM and OA, as well as productivity tools like Meitu. A Jefferies report points out that industry know-how is a key element differentiating WorkBuddy's platform, with its Agent runtime serving as the execution engine, integrating expert capabilities, skill modules, connectivity interfaces, and application scenarios.

At the hardware ecosystem level, WorkBuddy has partnered with over 30 hardware brands, covering categories such as smart glasses, robots, microphones, and recording equipment, and enables seamless cross-device access through cloud capabilities. The report emphasizes that large models play a crucial role in long-cycle tasks, while cloud architecture provides the fundamental support for multi-device collaboration.

Harness: Core Assets Driven by the Flywheel Effect

The Jefferies report lists Harness as Tencent's most strategically valuable asset and identifies several positive factors.

WorkBuddy boasts a significant scale advantage. It has accumulated over 70,000 skills on Harness, far exceeding the 5,000 to 10,000 skills of its competitors, creating a clear ecosystem barrier. Furthermore, WorkBuddy is compatible with multiple model families, rather than belonging to a single AI lab model team, giving it a differentiated advantage in model flexibility.

The flywheel effect is beginning to emerge. The report points out that as the number of users and skills continues to grow, Harness is forming a positive flywheel: post-training feedback loops generated by user usage can feed back into model optimization, thereby improving the product experience and attracting more users and developers to join.

The monetization path is clear. Jefferies points out that Harness's current gross margin is similar to that of its GPU leasing business, but as free users convert to paying users, Harness's gross margin has significant room for improvement. Tencent has established a comprehensive KPI evaluation system that includes DAU, revenue per DAU, and token consumption per DAU, indicating that its management of Harness's commercialization process is becoming more refined.

In addition, WorkBuddy uses its self-developed Hy3 Production model as the default model for users, creating internal synergy in terms of cost-effectiveness and further strengthening Harness's competitive moat.

Risk Warning and DisclaimerInvesting involves risk; please exercise caution. This article does not constitute personal investment advice and does not take into account the specific investment objectives, financial situation, or needs of individual users. Users should consider whether any opinions, views, or conclusions in this article are suitable for their specific circumstances. Any investment decisions made based on this information are at your own risk.