1.2 billion yuan. The largest angel round of financing for nuclear fusion is here.
This year, the record for nuclear fusion financing has been broken again.
On April 10th, Nova Fusion announced the completion of a 700 million yuan Angel + round of financing. Looking back at its establishment pace, this company founded in April 2025 completed two rounds of financing in just one year: it secured 500 million yuan in the Angel round in August of the same year, and the cumulative financing has reached 1.2 billion yuan so far, breaking the financing scale and speed record of domestic private nuclear fusion enterprises in the early stage of establishment.
Since the beginning of this year, nuclear fusion financing has witnessed a dense explosion. In January, Starring Energy completed a 1 billion yuan Series A financing, setting a new high for a single - round financing; in February, China General Nuclear Power Corporation, in cooperation with Henan state - owned assets, promoted Zhongke Qingneng to complete nearly 500 million yuan in financing; thereafter, projects such as Yanchao Jueneng and Chaoci Xinneng successively completed tens of millions of yuan in Angel rounds. Meanwhile, the sources of capital have become more diverse, attracting a batch of cross - border funds from Alibaba, Meituan, NIO, Mihoyo, etc.
With the surge in computing power demand, the power constraint has changed from a “hidden bottleneck” to an “explicit crisis”, and nuclear fusion has been pushed back to the center of the stage. The 14th Five - Year Plan clearly includes nuclear fusion energy in future industries. In January this year, the “Atomic Energy Law of the People's Republic of China” officially came into effect, clearly stating “encouraging and supporting controlled thermonuclear fusion”.
Supported by policies, state - owned assets represented by energy central enterprises and local guiding funds, the main force in the industry's development, have begun to make arrangements. However, different from the “national team” which emphasizes project stability and long - term strategy, private enterprises are more concerned about achieving commercialization as soon as possible. This is where the appeal of Nova Fusion lies. It has chosen the FRC (Field - Reversed Configuration) small modular technology route, attempting to promote the evolution of fusion devices towards a distributed energy form with a simpler structure and lower engineering complexity, and deeply integrate with artificial intelligence scenarios, putting forward the vision of “Fusion for AI & Beyond”.
However, there is still a gap between the ideal and the reality. Despite the repeated record - breaking financing and the diverse technological paths, there are still very few projects in the current nuclear fusion industry that have truly achieved a commercial closed - loop. Most enterprises are still in the R & D and engineering stages with heavy investment and long cycles, and the first to run through the business model are not the companies building “reactors”, but the “shovel sellers” that provide equipment, materials, and engineering capabilities for reactors.
Entrepreneurship of a post - 1965 scientist
Before founding Nova Fusion, Guo Houyang had held key technical positions in several domestic private nuclear fusion enterprises.
In 2019, when ENN Group launched a compact fusion experiment, Guo Houyang served as the Fusion CTO; then he continued to lead the plasma physics R & D at Energy Singularity. In 2025, he officially founded Nova Fusion, choosing FRC as the core route, hoping to rapidly promote industrialization with a fusion system that has a simpler structure and lower engineering difficulty. Guo Houyang also once served as the director of the well - known EAST Tokamak Research Laboratory in China.
Guo Houyang is the pioneer inventor of the FRC - SMR modular technology, with profound experience in both the FRC and Tokamak routes. He once led the Sino - US cooperation in magnetic confinement nuclear fusion, served as an expert on the ITER International Science and Technology Advisory Committee, a member of the Magnetic Confinement Expert Group of the Ministry of Science and Technology of China, and the chief of a major fusion project, reaching the international top level in the field of fusion.
He was born in Gushi County, Henan Province in 1965. He graduated from Hefei University of Technology with a bachelor's degree, then entered the Institute of Plasma Physics of the Chinese Academy of Sciences and obtained a master's degree in 1988. After graduation, Guo Houyang went to the Institute of Energy and Materials of the University of Quebec in Canada to study abroad, obtained a doctor's degree in 1993, and was employed by the Canadian Nuclear Fusion Research Center.
Since then, Guo Houyang's career has spanned the largest fusion research institutions on three continents, becoming one of the very few scientists with profound attainments and leadership experience in both the Tokamak and FRC main technical routes.
Later, Guo Houyang served as the experiment leader of the world's largest Tokamak device JET (Joint European Torus) in the UK, leading cutting - edge physics experiments. Since 1999, he has been the chief research scientist at the University of Washington in the United States and the general leader of the FRC experiment; after many years, he joined the US company TAE, serving as the chief scientist, chief experimental strategist, and director of the experimental operation department, fully responsible for the FRC fusion R & D project. Later, he participated in the establishment of the US national - level DIII - D Tokamak Boundary Plasma and Materials Research Center and served as the director of this research center.
Currently, there are mainly three technical routes for controllable nuclear fusion: gravitational field - confined nuclear fusion, laser inertial confinement nuclear fusion, and magnetic confinement nuclear fusion. Among them, magnetic confinement nuclear fusion currently occupies the mainstream position, and the main technologies include Tokamak (toroidal magnetic field), linear field - reversed configuration (Field - Reversed Configuration, FRC), stellarator (complex external coil magnetic field), reversed - field pinch, and magnetic mirror.
Among them, the FRC - SMR technology route is a mature route on the verge of commercialization. As a representative enterprise of the FRC technology route, Helion Energy in the United States has set a plasma temperature record of 150 million degrees Celsius. Its first fusion power plant “Orion” under construction is planned to supply 50 MW of electricity to Microsoft's AI supercomputing center in 2028, and it has signed a power purchase agreement of at least 50 megawatts. At the same time, OpenAI also plans to purchase electricity from Helion, with a scale of up to 50 gigawatts.
The largest Angel round of nuclear fusion this year
Just one year after its establishment, Nova Fusion has gathered a group of leading institutions and industrial capital, reflecting the rare capital enthusiasm in the current nuclear fusion track.
Going back to August 2025, the company completed a 500 million yuan Angel round of financing. The list of investors included the Zhongguancun Independent Innovation Special Fund of the National Social Security Fund (managed by Legend Capital), Legend Capital, Photosynthetic Venture Capital, Gao Rong Venture Capital, Huakong Fund, Ming Shi Venture Capital, Yunqi Venture Capital, and Lingang Science and Technology Innovation Investment. In the recently completed 700 million yuan Angel + round of financing, the capital structure has been further “upgraded”, forming a diversified combination of “Internet industrial capital + leading financial investors + state - owned innovation funds”. Among them, Alibaba, as an old shareholder, chose to increase its investment, and Meituan Longzhu extended its reach to the nuclear fusion track for the first time.
“At this stage, money is chasing and betting on technology. But for entrepreneurs, 2026 is the best window period. If you don't take action now, when will you?” The chairman of a second - level subsidiary of a science and technology central enterprise said in an interview with the media about the current industry boom.
At the beginning of 2026, Starring Energy completed a 1 billion yuan financing, breaking the single - round financing record of private nuclear fusion. In March, “Yanchao Jueneng”, a representative enterprise of the stellarator route in China, completed a tens of millions of yuan Angel + round of financing, jointly invested by Honglida, Hongyuan Investment, etc. The company is about to launch a Pre - A round, and more than 50% of the quota has been subscribed. This is a rare financing case of a non - Tokamak mainstream route in China.
Meanwhile, the “national team” is also entering the arena. The National Energy Administration has established a 20 billion yuan fusion industry fund, and the Shanghai Future Industry Fund has been expanded to 15 billion yuan. Long - term funds at the state - owned asset level are accelerating their entry. Chen Zhongyong, the first - batch expert of the ITER program of the Ministry of Science and Technology, also gave a set of quite impactful comparison data to the media: in the first half of 2025, the total financing of domestic private nuclear fusion has exceeded 11.5 billion yuan, while before 2019, this figure was almost zero.
From a global perspective, this upsurge is also clear. The US nuclear fusion startup Inertia Enterprises recently completed a $450 million Series A financing to promote the development of a laser fusion power plant and plans to achieve commercialization around 2030. According to data from the US Nuclear Fusion Industry Association, as of mid - 2025, the cumulative global nuclear fusion financing has exceeded $9.7 billion, entering an accelerated climbing stage.
And the underlying logic behind all the capital influx is being strengthened by a more certain variable, that is, the explosion of energy demand, especially the “devouring” of electricity by AI. Huaxi Securities estimates that the annual electricity consumption of global data centers is currently about 415 terawatt - hours, and it is expected to jump to 945 terawatt - hours by 2030, almost doubling.
“From a physical path perspective, human energy utilization at the electronic level is approaching its limit, and the next step can only be to move towards energy release at the atomic nucleus level.” Mi Lei, the founding partner of Zhongke Chuangxing, said bluntly.
Currently, it's still the “shovel sellers” who make money
The commercial prospects of nuclear fusion are beyond doubt. According to Goldman Sachs research, it is estimated that by 2030, the global data center's electricity demand will reach 1130 TWh, equivalent to Japan's annual electricity consumption, while traditional energy cannot meet the high - power density and 7x24 - hour uninterrupted power supply requirements of AI data centers. Altman, the founder of OpenAI, has mentioned nuclear fusion many times in public, believing that AI and nuclear energy are closely related.
However, controllable nuclear fusion is a typical “time - for - space” industry. It has extremely high technical thresholds, huge engineering complexity, and a long commercialization cycle.
Currently, the industry is standing at a critical threshold. On the one hand, the competition among different technical routes continues. Represented by the “national team”, they are more inclined to the low - temperature superconducting system, emphasizing project stability and long - term controllability; while some private enterprises, among others, are betting on high - temperature superconductivity, hoping to achieve device miniaturization and accelerate commercialization through a higher magnetic field strength.
2027 is generally regarded as the first “watershed” for the two paths. A person from a nuclear fusion enterprise in East China said frankly: “If the ‘national team’ clearly bets on low - temperature superconductivity after 2027, our enterprises that bet on high - temperature superconductivity may face the risk of a contraction in policy support, and the investment of billions of yuan in the early stage is uncertain.”
On the other hand, even if the technical direction becomes gradually clear, it still takes time to achieve stable operation with a “Q value > 1 (output energy greater than input energy)” and continuous power generation. The improvement of policies, regulations, and safety systems will not happen overnight either.
In Guo Houyang's view, the future of China's fusion industry is not a zero - sum game between the “national team” and “private forces”, but a coordinated operation between “cornerstones” and “shock troops”. That is, the national team focuses on basic scientific research and fundamental technology R & D, aiming at large - scale centralized base - load energy after 2040; Nova Fusion, with the FRC - SMR route as the core, precisely captures the distributed energy gap brought by AI computing power. With its technological first - mover advantage, top - notch international team, and abundant capital, it can iterate rapidly with the efficiency of a startup and be the first to fill the strategic gap in the short - term demand for clean energy.
However, the differentiation of technical paths does not affect the certainty of upstream revenues. In the Chinese A - share market, a group of “hidden champions” focusing on segmented segments have begun to fulfill orders and achieve performance. For example, Hefei Forging Press Intelligent Equipment Co., Ltd., relying on its accumulation in the high - end equipment manufacturing field, has entered the manufacturing of core components of Tokamak and obtained substantial orders; as the tender scale of NeoFusion exceeded 800 million yuan in the first quarter of 2026, as a key part of the supply chain, it is expected to directly benefit from the industry's growth. Another example is Baili Electric. Its subsidiary has made a breakthrough in the field of high - temperature - resistant electromagnetic wires, and the relevant products have been delivered, which is one of the key materials for high - temperature superconducting magnet manufacturing; another subsidiary has participated in the construction of power system supporting equipment for the ITER project.
This means that on the way to the “ultimate energy”, although capital has flocked in, at present, it is still those who sell tools that make money.
This article is from the WeChat official account “Dongshisi Tiao Capital” (ID: DsstCapital), author: Wei Xianghui, published by 36Kr with authorization.