一个世界级芯片团队落地杭州

This article centers on Zhaoguan Electronics, a world-leading visual processing chip (VPU) company, which will relocate its headquarters to Linping, Hangzhou by the end of April 2026. The move is a typical case of collaborative AI industry development among startups, investment institutions and local governments. Key takeaways are as follows:

1. Team and business: Founder Xinpeng Feng was the youngest global R&D director for chips in AMD’s history, while co-founder Ji Zhou previously served as AMD’s chief scientist. The team has led the design of more than 50 CPU/GPU products. Founded in 2016 after returning to China, the company focuses on developing VPU for visual intelligence.

2. Market positioning: The second wave of AI development has entered the "Physical AI" era. VPU serves as the "eyes" of Physical AI, enabling real-time on-device processing of massive visual data to meet the low-latency, high-reliability requirements of use cases such as robotics. Zhaoguan’s technology can compete with market-dominant Intel, and its products already serve more than 200 downstream scenarios across 8 major industries.

3. Local industry impact: The relocation fills a key gap in Hangzhou’s AI and visual intelligence industry chain. Leveraging its geographic advantages and mature industrial supporting facilities, Linping is well-positioned to nurture a 100-billion-yuan-level leading enterprise from this investment.

This article offers valuable insights for brand owners planning to enter AI-related businesses, helping them capture industry trends and access development resources. Key takeaways are as follows:

1. Industry and consumer trends: AI development has shifted from the digital world to the physical world. Physical AI and embodied intelligence represent a trillion-dollar market, where VPU for visual intelligence is a core and essential component. Its downstream applications cover many high-growth scenarios including humanoid robots, industrial robots, drones and low-altitude economy, with enormous market potential.

2. Product R&D direction: On-device Physical AI has rigid requirements for low-latency, high-reliability visual processing. The global VPU market has long been dominated by Intel, leaving enormous room for domestic substitution. A full-stack self-developed, on-device focused technology strategy better aligns with market demand and delivers stronger competitiveness.

3. Local resource reference: Located at the intersection of three major technological innovation corridors in the Yangtze River Delta, Linping, Hangzhou is home to China’s first computing power town that serves as a solid digital base. It offers AI industry policies aligned with national strategic plans and a mature collaborative investment-government investment promotion mechanism, making it an ideal location for hard technology brands to establish operations.

This article provides clear opportunity assessment and risk reference for sellers targeting the AI track. Key takeaways are as follows:

1. Policy and market opportunities: China’s 15th Five-Year Plan has explicitly identified AI terminals, embodied intelligent robots and low-altitude economy as key development sectors. Hangzhou has built a "296X" advanced manufacturing cluster system to foster a complete AI ecosystem spanning from chips to end applications. Sellers in related downstream sectors can leverage policy dividends to expand their market footprint.

2. Supply chain opportunities: While the VPU market was previously monopolized by Intel, China now has domestically grown players such as Zhaoguan Electronics with world-class technology. Sellers operating in robotics, drones and other related downstream sectors can partner with domestic suppliers to cut costs and reduce exposure to risks from overseas monopoly.

3. Growth opportunity note: Linping, Hangzhou has a mature collaborative investment-government investment promotion model and has attracted a large number of high-tech enterprises with complete supporting ecosystems. AI-related sellers can settle locally to achieve faster growth by leveraging the ecosystem. At the same time, players should note that the AI sector features rapid technological iteration, so they need to keep up with technology trends and make layouts in advance.

This article offers clear reference for commercial opportunities and transformation directions for factories that serve the AI industry. Key takeaways are as follows:

1. Business collaboration opportunities: The Physical AI track is currently entering a period of rapid growth, driving surging demand for VPU. As a leading domestic VPU enterprise, Zhaoguan Electronics’ products cover more than 200 downstream scenarios, and its business expansion requires stable manufacturing support. Domestic factories engaged in wafer manufacturing, chip packaging and testing and other related fields can pursue collaboration opportunities to access the high-end chip supply chain.

2. New business opportunities: The rapid growth of Physical AI has driven up demand for AI terminal hardware including various robots and drones. Related terminal manufacturing factories can expand production capacity for these fields to capture growth dividends from the trillion-dollar track.

3. Insights for digital transformation: The development of hard technology increasingly relies on local industrial ecosystems and digital infrastructure. When advancing digital transformation, factories can proactively connect with local policies and computing power resources, accelerate upgrading through policy dividends, integrate into the AI industry supply chain, and improve their own competitiveness.

This article sorts out industry trends and market demand for service providers related to the AI industry. Key takeaways are as follows:

1. Industry development trends: AI development has entered its second phase, with the focus shifting from the digital world to the physical world. Physical AI and embodied intelligence have become a new trillion-dollar track, driving demand for a range of related services including chip design services, supply chain supporting services, computing power services and investment matchmaking services, with large market gaps and broad growth space.

2. New technology service demand: On-device VPU is a core necessity for Physical AI, and China’s VPU track is currently accelerating domestic substitution, which requires a large amount of supporting services including technical R&D services, intellectual property services and supply chain supporting services. Related service providers can make targeted layouts to capture market share.

3. Customer pain points and solutions: When expanding and relocating, hard technology chip enterprises’ core demand is access to local government resources and industrial capital. Service providers can build matchmaking platforms connecting investment institutions, local governments and technology enterprises, and develop tripartite collaborative investment promotion solutions to meet the core demand of enterprises for site expansion.

This article provides strong reference for the operation and development of industrial platforms and investment promotion platforms. Key takeaways are as follows:

1. Core demands of enterprises on platforms: When hard technology chip enterprises reach the key expansion stage, their core demands are complete industrial chain supporting facilities, clear industrial policy support and smooth capital access channels. When selecting locations, they value the platform’s geographic advantages and ecosystem integrity more than standalone preferential policies.

2. Replicable innovative investment promotion model: Linping, Hangzhou adopts a collaborative investment promotion model led by leading investment institutions and local governments. Leveraging investment institutions’ industry expertise and project pipeline, the model accurately attracts high-quality projects that align with local industrial planning. It has successfully brought in multiple high-tech projects and is mature and replicable.

3. Operation management insights: Platforms should target gap-filling projects in line with their own industrial chain planning. Zhaoguan Electronics’ relocation fills a key gap in Hangzhou’s visual intelligence industry and helps improve the city’s overall AI ecosystem. Platforms should avoid blind investment attraction, make layouts based on their own advantages, and mitigate the risk of project loss caused by incomplete industrial chains.

This article provides the latest industry case and research materials for AI chip industry research. Key takeaways are as follows:

1. New industry development trends: AI development has entered its second phase, and Physical AI and embodied intelligence have now become a global industry consensus. As the core entry point connecting Physical AI to computing systems, VPU is accelerating domestic substitution. China has already seen domestic players capable of competing with global giants, and the competitive landscape of the global VPU track is undergoing shifts.

2. Innovative industrial cooperation model: This article presents a new development model for hard technology: in-depth collaboration among entrepreneurial technical teams, leading specialized investment institutions and local governments. Investment institutions not only invest in projects, but also leverage their resources to help projects connect with local sites. Local governments gain high-quality projects to complete their industrial chains, while enterprises gain development resources, achieving a triple win.

3. Insights for industrial policy: For local governments developing hard tech industries, it is critical to improve infrastructure based on geographic advantages, proactively address gaps in industrial chain planning, and attract leading enterprises in key segments to drive the development of entire industrial clusters. This model offers strong insights and reference for AI hard tech industry development across China.