Funding Announcement

On May 7, Wenxin Qionggong announced that it has secured over 700 million in funding, maintaining its position as the leading AI-native infrastructure company in China. The round was co-led by Hangzhou High-tech Jin Investment Group and Huiyuan Capital, with participation from Guoxing Capital, Qinhuai Data, GF Qianhe, Lihe Qingtong, Zhongbao Investment, AEF NextGen, Tengrui Capital, Kalete, CITIC Jinzheng Capital, and Kuande Intelligent Learning Laboratory, along with additional investments from existing shareholders Junlian Capital, Shanghai Guotou Futeng, and Yuanzhi Future.

This financing round was completed several months ago and will focus on three main areas: strengthening Wenxin Qionggong’s leading position in diverse heterogeneous technologies, expanding the Token economy, and optimizing usable computing power; enhancing the technical advantages of software-hardware synergy to accelerate the leap from “electricity to Token” in production efficiency; and building AI infrastructure with self-evolution capabilities, providing solutions tailored to various industry scenarios through an enterprise-level intelligent service platform to enhance the value conversion from “Token to productivity.”

Token Economy Era

This funding reflects a deep recognition from investors of the core value of AGI infrastructure in the Token economy era. As the global industry transitions from “model competition” to “AI productivity,” Wenxin Qionggong has officially released its AI productivity formula: “AI Productivity = Intelligent Scale × Token Production Efficiency × Token Value Conversion.” This formula quantifies the process of releasing productivity value in the Token economy era and, leveraging its unique “high efficiency + large scale + intelligent” Agentic Infra system, the business is experiencing rapid growth, becoming a hub for China’s Token economy.

Wenxin Qionggong’s co-founder and CEO, Xia Lixue, stated: “In the Token economy era, AGI infrastructure plays a role similar to that of a refinery in the petrochemical industry, converting energy into digital oil (Token) and providing essential resources for various AI terminal applications and services. The AI productivity formula we proposed is our systematic answer to how to build a high-efficiency, large-scale, and intelligent Agentic Infra. The recent funding will accelerate our comprehensive breakthroughs in these three dimensions, responding to the core issues of the global Token economy with a Chinese solution.”

Growth in Token Demand

Recently, the Token economy industry driven by Agentic AI models has exploded. Wenxin Qionggong has not only built a solid AI-native technology foundation through software-hardware synergy and diverse heterogeneous technological advantages but has also keenly captured the trends of multi-agent collaboration and end-to-end implementation, leading the construction of a new generation of self-evolving Agentic Infra.

As of the end of April this year, the daily Token call volume of Wenxin Qionggong’s Agentic MaaS large model service platform has increased over 20 times compared to the end of last year, continuously adapting open-source models for Day0 deployment. The latest data released by the National Bureau of Statistics shows that as of March this year, China’s daily Token call volume has surpassed 140 trillion, a growth of over 40% compared to the end of last year. AI is deeply empowering various industries, driving rapid expansion of the digital economy. This explosive growth is reminiscent of the early stages of mobile data in the 3G era—back then, 100MB of monthly data seemed trivial, and today’s Token usage also has exponential growth potential.

The global demand for Tokens will exponentially leap along the path from human users to coexistence with intelligent agents and finally to dominance by intelligent agents. It is expected that after 2027, the scale of Tokens consumed autonomously by machines will far exceed that consumed by humans, driving a massive explosion in global Token demand.

The long-term value of AGI infrastructure lies not only in technological leadership but also in promoting the construction of a sustainable Token economy ecosystem.

Chinese Characteristics in Token Economics

In March of this year, Xia Lixue first proposed the concept of “Chinese Characteristics Token Economics” at the 2026 Zhongguancun Forum: to connect the complete economic chain from “investment → electricity → Token → productivity → value,” allowing Tokens to upgrade from a technical measurement unit to a core economic variable, truly achieving sustainable development of the AI industry.

China possesses abundant energy structure advantages, a complete AI industry chain, and the world’s largest AI application market, fully capable of replicating the successful path of “Made in China.” With high efficiency, large scale, and intelligent Agentic Infra, Wenxin Qionggong transforms resource advantages into high-quality AI productivity, injecting Chinese momentum into global AI development.

Facing this historic industrial opportunity, Wenxin Qionggong has officially released the AI productivity formula, deeply anchoring the Token economy with the company’s core capabilities. It also aims to provide new insights into the question of “how AGI infrastructure can enhance the productivity of society as a whole.”

AI Productivity Formula

The AI productivity formula proposed by Wenxin Qionggong is: “AI Productivity = Intelligent Scale × Token Production Efficiency × Token Value Conversion.”

• Intelligent Scale: Achieved through the extreme optimization of diverse heterogeneous computing power. During the explosive period of the Token economy, computing power scale has become a core bottleneck for intelligent limits. The supply and cost structure of computing power from a single chip manufacturer is restricting the process of scaling AI implementation. Wenxin Qionggong utilizes “diverse heterogeneous” core technology to achieve extreme optimization across various model algorithms and chip hardware, significantly enhancing the usable scale of computing power and breaking the current computing power shortage dilemma.

• Token Production Efficiency: Producing more Tokens from the same resources, measured in “Token/s.” Wenxin Qionggong has launched a large-scale model training and reasoning integration platform for AI-native enterprises through its Agentic Infra autonomous AI foundation, optimizing through software-hardware synergy and deeply optimizing and jointly designing across software and hardware to maximize chip application computing power.

• Token Value Conversion: Creating greater productivity from the same Tokens, measured in “productivity/Token.” The Wenxin Qionggong platform has integrated leading large models, including Kimi, Zhipu, DeepSeek, Tongyi Qianwen, and MiniMax, further optimizing and maximizing the intelligent limits of trillion-parameter open-source large models, providing system-level Token productivity services to AI-native enterprises, manufacturing, cultural and film industries, and smart terminals.

The multiplication of “Token Production Efficiency” and “Token Value Conversion” yields “productivity/s”—throughout human industrial civilization, the industrial revolution and the invention of the internet have pointed towards a leap in “productivity per second” for society. Currently, Tokens have become a new unit of productivity value, internalized in the process of maximizing productivity value release.

Agentic MaaS Large Model Service Platform

As one of the company’s core products, the Wenxin Qionggong Agentic MaaS platform employs a multi-layer architecture design, providing full-stack system capabilities from model access, traffic governance, reasoning optimization to resource scheduling for large model companies and AIGC application companies needing to access large model services. It deeply applies intelligent agent capabilities, dynamically adjusting underlying model resources based on service performance and load conditions, ensuring 24/7 high availability.

The platform provides high-performance service optimization for mainstream domestic open-source models such as Zhipu, Kimi, DeepSeek, Tongyi Qianwen, and MiniMax, and has conducted deep reasoning optimizations for trillion-parameter large models.

In real production environments, the Agentic MaaS platform demonstrates several industry-leading performance advantages. In terms of execution accuracy, the platform can support complex toolchains while ensuring a high degree of consistency with the original model’s performance (accuracy alignment rate >99.9%). In terms of operational efficiency, the platform has significantly increased system throughput by 2 to 3 times, reduced overall latency by 50%, and controlled the first-word delay within 500ms. Additionally, an enterprise-level high availability rate of 99.95% provides solid stability support for various real business operations.

Currently, the Agentic MaaS platform has launched over 160 large models, all of which support plug-and-play. As of the end of April, the daily Token call volume of the Agentic MaaS platform has increased over 20 times compared to the end of last year, building a strong growth flywheel of “continuously lowering reasoning costs → deep cooperation with leading large model companies → accumulation of extensive real business experience → rapid iteration of reasoning optimization technology → continuous improvement of performance and reliability → rapid increase in daily call volume.”

Investor Responses

The head of the co-leading investor Hangzhou High-tech Jin Investment Group stated: “High-tech Jin Investment Group is committed to empowering scientific innovation with capital and serving the real economy, focusing on national strategic frontiers such as artificial intelligence, integrated circuits, and the digital economy, cultivating independent innovation capabilities. Leading this round of investment in Wenxin Qionggong is not only a high recognition of the team’s Tsinghua gene, technical depth, and industrial implementation capabilities but also a long-term confidence in China’s AGI new infrastructure being autonomous and controllable. We will leverage the high-tech zone’s industrial ecosystem, policy resources, and post-investment empowerment system to provide full-chain support of capital + scenarios + ecology, helping Wenxin Qionggong continue to lead in the AGI infrastructure track and accelerate the transformation and large-scale application of technological achievements.”

Zhang Shanshan, president of co-leading investor Huiyuan Capital, expressed: “AGI infrastructure is the core industrial foundation of the Token economy era, and its strategic value is comparable to the fundamental transformations brought by electricity and the internet to overall social productivity. As a co-leading investor, Huiyuan Capital will leverage its AI ecosystem accumulation in data centers and industrial application scenarios to fully utilize the advantages of industry-investment integration, empowering corporate growth and helping Wenxin Qionggong accelerate the rapid expansion of the Token economy hub. We firmly believe that Wenxin Qionggong will become a benchmark force in China’s AGI infrastructure.”

He Dingkang, head of the AEF NextGen Fund, the first overseas investor in Wenxin Qionggong, stated: “We are honored to participate in this round of financing. Under the excellent leadership of its management team, Wenxin Qionggong is playing a key role in continuously driving innovation and development in the Token economy in China and globally. With the resources and network of AEF NextGen Fund in Hong Kong and overseas, we will continue to deepen cooperation with Wenxin Qionggong to jointly explore business opportunities in Hong Kong and international markets.”