Market interest in securing a stake in DeepSeek has reached unprecedented levels, with investors employing various methods to obtain an allocation. Investor Li Jing, acting on behalf of others to find an allocation, made three trips to DeepSeek's headquarters in Hangzhou, waiting outside each time, but was unable to meet founder Liang Wenfeng in person. In late May, an intermediary claiming to have connections to Liang proposed a 5 million yuan "meeting fee" to arrange an introduction. Li Jing paid the fee and arrived in Hangzhou two days early, but later abandoned the meeting after hearing the financing situation had changed and successfully requested a refund. Separately, a state-owned fund investor contacted eight financial advisors (FAs) who each claimed to have allocations, with almost none of the leads materializing.
Key Details of the Funding Round
DeepSeek, which had previously insisted on not raising external capital or pursuing an IPO, has recently completed its first funding round exceeding 50 billion yuan. The process was marked by investors scrambling for access and intermediaries aggressively marketing allocations, creating a mix of facts and chaos. A person involved in the transaction revealed that the AI foundation model company DeepSeek has concluded its first-round financing talks, with agreements signed at the end of last month. Media reports indicate this round raised over 50 billion yuan, approximately $7.4 billion. Within this, Liang Wenfeng personally contributed 20 billion yuan, the largest single investment in the round. Tencent invested 10 billion yuan, CATL invested 5 billion yuan, while NetEase, JD.com, and IDG Capital each invested 3 billion yuan. The National AI Industry Investment Fund contributed 1 billion yuan.
The source added that following this round, DeepSeek's pre-money valuation is about 350 billion yuan. However, after accounting for a 5% increase in the ESOP (employee stock ownership plan), the actual pre-money valuation is approximately 367.5 billion yuan, or about $54.3 billion. The final list of investors shows significant changes from earlier market rumors, with several leading corporate strategic investors and market-oriented investment institutions entering the fray. Notably, what surprised the transaction participant was that the allocation for state-owned capital ultimately turned out to be the smallest.
Since April, sentiment in the primary market has been driven by DeepSeek's continuously rising valuation, which climbed from $20 billion in April to approximately $54 billion currently. Upon completion of this funding round, DeepSeek will become the company with the largest first-round financing in the history of China's AI large model sector. This marks the first time the three-year-old company has opened to external investment, having previously adhered to a founder-proclaimed stance of "no financing, no IPO, no commercialization."
Intermediaries and Questionable Allocations
Many investment institutions have been actively seeking channels, leading to the emergence of various brokers and middlemen of unknown origin. Investor Li Jing was also tasked by others to find an allocation. After news of the financing emerged, he traveled to Hangzhou three times specifically to meet Liang Wenfeng, but returned empty-handed each time. In late May, Li Jing was contacted by an introducer claiming proximity to Liang, who proposed a 5 million yuan "meeting fee" to arrange a meeting. Li Jing arrived in Hangzhou two days early, staking out the area near DeepSeek's headquarters. Later, upon hearing the financing situation had changed, he abandoned the meeting and successfully requested the 5 million yuan back from the introducer.
"The ability to pay is itself a capability," Li Jing said. For the vast majority of LPs (Limited Partners), the most challenging issue is not how much to invest, but rather the inability to find the right person to give the money to. With opaque channels, people are already profiting at every intermediate step. DeepSeek is undoubtedly one of the hottest AI investment targets domestically. Despite the first-round investor list being finalized, social media is still flooded with posts offering DeepSeek allocations.
In recent years, the global AI investment boom has continued, driving up valuations for hot targets. After U.S. large model startup Anthropic completed its latest $6.5 billion funding round, its valuation rose to $965 billion, surpassing OpenAI to become the world's highest-valued AI unicorn. For hot targets like Anthropic, investing through an SPV (Special Purpose Vehicle) was a common industry practice, where some GPs (General Partners) package assets into tradable securities to bring in new LPs, who often lack the name recognition for direct investment.
Pass-through investing via a special purpose fund incurs additional management fees, a method less commonly used by top-tier funds to enter a financing round. However, many investors stated that even with extra channel fees, they were willing to invest, largely due to a prevailing belief that DeepSeek's valuation will continue to rise, promising profits upon a future IPO.
When inquiring with one channel contact while posing as an LP, they claimed the current channel secured a 5 billion yuan allocation from the "National Integrated Circuit Fund," limited to individual participation. Investors would enter as primary LPs directly into the state-owned shareholder structure, were required to submit sufficient proof of assets within two days, with subscription expected to be completed within the week. They also mentioned that the FA and GP's upfront management fees for this transaction were 8% and 9% respectively, totaling a hefty 17% upfront cost. This is highly unusual in standard private equity transactions, where GPs typically charge an annual management fee of around 2% of the fund size. It is understood that intermediaries selling DeepSeek allocations often propose collecting management fees for 3 to 5 years upfront.
These FAs also create an atmosphere of extreme scarcity. Another FA, claiming to have secured a 20 billion yuan DeepSeek allocation from a "local state-owned entity," stated they had several channels, with a 20 billion yuan fund typically "sold out in two to three days." However, when asked for the specific GP's name, they became evasive, only mentioning a "long-term cooperative relationship." Yet, half a month later, upon follow-up, the same intermediary indicated that the same channel still had available quota.
Despite unclear source channels, some are still willing to take the risk. Investor Chen Duo, serving at a state-owned fund managing tens of billions, negotiated with eight FAs in total. These FAs claimed allocations came from "certain local state-owned entities," but in most cases, communication ceased after Chen Duo sent a POF (Proof of Funds). Chen Duo stated that any claim unable to provide on-site verification of the underlying investment agreement or confirmation letter was considered unreliable.
Through these interactions, Chen Duo gradually understood the fundraising logic: many institutions first raise a fund to secure capital, then use the money in their account to request an allocation from DeepSeek. FAs then create a seemingly reliable channel using rhetoric like connections to senior officials, urging investors to quickly provide proof of funds. "These institutions want Liang Wenfeng to think they already have money in their account to get the allocation." This creates a chicken-and-egg cycle, as even with sufficient raised capital, securing a DeepSeek allocation is not guaranteed.
An investment manager at a local state-owned fund expressed frustration that FAs still approach him to introduce LPs, stating, "DeepSeek simply doesn't lack LPs and doesn't need FAs." He believes ordinary people wouldn't trust these unclear channels, unless they are very small institutions or individual LPs who might be persuaded.
A person familiar with DeepSeek's dealings revealed that all capital entering this round came from main funds, with no allocations through any special purpose funds. Media reports also indicated that Liang Wenfeng and his team required verification of all LP identities participating in this round to prevent allocations from falling into the hands of unknown investors. The source stated that FAs still fundraising ultimately won't succeed, questioning their motives.
The Selection Process for Investors
Tang Yu, a partner at a private equity fund, has been following DeepSeek since its rise in February 2025. After news of the open financing emerged, Tang Yu, privately commissioned by a state-backed fund with 3 billion yuan investment intent, sought channels everywhere. Through personal connections, he contacted a senior executive at High-Flyer Quant, a quantitative hedge fund controlled by Liang Wenfeng that has long funded DeepSeek's R&D costs.
Despite years of acquaintance, Tang Yu found that his 3 billion yuan offer did not open doors as expected. The executive only provided a company email, instructing him to send institutional introductions, letters of intent, and cooperation proposals for "unified review by responsible personnel," after which he received no further positive response. Tang Yu noted that all interested parties able to contact this executive likely received similar replies. On May 28, when he inquired again about financing developments, the response was that participating in this round was certainly too late, with the timing of the next round unknown.
Multiple investors involved in the transaction stated that Liang Wenfeng personally selected investors for this round, preferring institutions that wish for his success or have good relations with him. Contact was mostly initiated by Liang's invitation; those not invited found it difficult to establish truly effective connections, with repeated meeting attempts failing to yield cooperation. Another investor, Hu Jie, also noted that funds unfamiliar with Liang or lacking name recognition basically couldn't get in.
Hu Jie works at a market-oriented investment institution where AI is a key focus area. However, at the start of DeepSeek's open financing, his interest was limited, as initial understanding suggested the round might be state-capital-led, the deal size was already large, and the investment return might not be high. Several institutional investors shared this view, initially believing it was a state-capital-dominated deal.
However, after this year's May Day holiday, Hu Jie observed several investment institutions, including Monolith Capital and Hillhouse Capital, intensively engaging with DeepSeek, prompting his institution to join the negotiations. To the surprise of several transaction participants, the final state-owned capital allocation was lower than expected. Media reports indicate that in this round, only the National AI Industry Investment Fund invested directly in DeepSeek, free from lock-up restrictions and enjoying company voting rights. Other institutional investments were made into limited partnerships managed by Liang Wenfeng, not directly into DeepSeek itself, with shares locked up for five years during which they cannot be sold.
Public information shows the National AI Industry Investment Fund Partnership was established in early 2025 by Guozhitou Private Fund Management Co., Ltd. and the National Integrated Circuit Industry Investment Fund Phase III Co., Ltd. Guozhitou, founded in 2024, has focused its recent investments on AI, chips, and robotics, including companies like Moonshot AI, MetaX, and Leaper Robotics.
In this funding round, many state-backed funds actively sought allocations. Chen Duo, serving a state-owned fund, contacted eight FAs before reaching the periphery of the deal. Another investment manager at a local state-owned fund actively engaging with AI companies stated he began paying attention to DeepSeek during the 2025 Spring Festival, also eyeing other star model companies like KNOWLEDGE ATLAS and Moonshot AI, but failed to invest in any, even lacking access. "Objectively speaking, state-owned capital has its limitations," he said. "Large companies require rapid capital deployment with large single-ticket sizes. For state-owned capital, this implies risk."
The investor list also includes several major tech firms. Among them, Tencent secured the largest allocation, followed by CATL. Previous media reports suggested Alibaba was also in negotiations with DeepSeek, though Alibaba has publicly denied this, expressing limited interest.
Multiple AI investors analyzed that between Tencent and Alibaba, Tencent had stronger motivation to participate, as its foundational model capabilities have long lagged behind the first tier, creating a stronger desire for strategic cooperation with DeepSeek. Alibaba, having built its own ecosystem from models to chips, has less strategic urgency to invest in DeepSeek. A transaction participant also believed DeepSeek's cooperation basis with Alibaba was weaker than with Tencent.
CATL's participation as a new energy giant indicates AI model vendors are strategically considering the energy variable behind computing infrastructure. A transaction source revealed that DeepSeek is building its own data centers, planning to operate computing power directly. Public information shows that in late April, DeepSeek posted several computing-power-related job openings, including Senior Data Center Operations Engineer and Senior Delivery Manager, requiring management of the full lifecycle from project initiation to operation, with locations in Ulanqab.
Valuation Rationale for DeepSeek
According to multiple AI sector investors, DeepSeek's decision to open financing stems from at least two reasons. First, DeepSeek's model iteration requires continuous large-scale investment in training compute, with sufficient capital being crucial for sustaining this long-cycle effort, as self-generated assets from High-Flyer Quant may not meet future R&D needs.
Chen Shi, partner at Ruifeng Capital, analyzed that top large model companies like Anthropic operate the industry's "premium business model," which is also the "most pressured business model" because Scaling Law remains effective. Each generation's model training cost exceeds the previous, requiring continuous investment in compute and data; falling behind means being forgotten by the market. As an AI startup, today's primary market can hardly meet the funds needed for model iteration, leaving the secondary market as the next step.
Second, it allows employee stock options to be market-validated, helping retain talent. This round increased the employee option pool by 5%, worth approximately 17.5 billion yuan. However, previous analysis indicated that as of the V4 model release, DeepSeek's core research team had not experienced talent attrition issues.
Hu Jie noted that DeepSeek employees also hope for a company IPO. An AI practitioner familiar with DeepSeek's R&D staff said it's unsurprising employees expect an IPO, stating, "People aren't saints; those sticking with DeepSeek can't rely on ideals alone."
Since 2025, several domestic star AI large model companies have been intensively fundraising and pursuing listings, accelerating capitalization. Besides KNOWLEDGE ATLAS and MiniMax listing on HKEX earlier this year, StepFun reportedly raised about $2.5 billion in its latest round with an expected cornerstone valuation around $10 billion. Qichacha data shows Moonshot AI has raised approximately $4.4 billion cumulatively, with a latest pre-money valuation of $20 billion. Moonshot AI is dismantling its VIE and red-chip structure in preparation for listing, while StepFun aims to file as early as the first half of this year.
In contrast, DeepSeek started fundraising the latest, yet its first round reached about $7.4 billion with a valuation of approximately $54.3 billion, far surpassing its competitors. Both internal and external factors have driven DeepSeek's continuously rising valuation above other model companies. Multiple transaction participants consider the current range reasonable, as in terms of foundational model capability, DeepSeek's inference efficiency is recognized as industry-leading.
Since its breakthrough R1 model last year, DeepSeek has maintained low cost and high performance. Taking the V4-Pro released in April as an example, official disclosures show that in a 1 million Token long-context scenario, the FLOPs per Token inference is only 27% of the previous V3.2 generation. This efficiency advantage is directly reflected in API pricing, with V4-Pro at 0.025 yuan per million Tokens, among the lowest globally for mainstream models. This extreme low cost has made DeepSeek highly popular among global developers, ranking first in Token consumption this month on the AI model aggregation platform OpenRouter's latest chart.
A professional familiar with the AI sector at a USD fund noted that on the OpenRouter platform, Chinese model call volume has jumped from 1% three years ago to 60%. Chinese model companies have successfully extended the supply chain's extreme cost-reduction advantage to the AI field, gradually making large models a standardized commodity, with DeepSeek executing this most effectively.
Additionally, multiple investors and industry experts analyzed that another reason for DeepSeek's high valuation is its rising status as a national-level AI strategic platform. External factors also play a role; the secondary market performance of KNOWLEDGE ATLAS and MiniMax has stimulated pricing for later entrants. Taking KNOWLEDGE ATLAS as an example, its IPO market cap was 57.9 billion HKD, with the latest market cap rising to 740.1 billion HKD. A state-owned fund professional familiar with AI venture capital analyzed that after this year's Spring Festival,对标 model companies including KNOWLEDGE ATLAS and MiniMax rose rapidly, creating a strong capital market wealth effect.
This fund professional also believes the current moment is a key node for discussing large model commercialization, as after three to four years of development, technology can support mature product deployment, with the market eager to verify AI's commercial value. Against this backdrop, more primary market capital is flowing into large model companies, whereas in the past, these model startups were generally viewed by state-backed funds as targets "lacking certainty."
Contrary to the overall market's "fear of missing out" (FOMO) sentiment, some investors who never sought contact believe DeepSeek's current valuation is too high and "unfathomable," with negotiations consuming significant time and monetary cost, and the final probability of successful investment too low, making the transaction unattractive considering investment returns. Hu Jie views DeepSeek as a "large white horse," referring to high-market-cap, relatively stable "top performers."
Facing capital market tests, DeepSeek cannot avoid issues all tech startups face: how technological ideals and commercial returns can coexist long-term. How to commercialize is a crucial question hanging over DeepSeek's high valuation. According to previous reports, in March, DeepSeek executives revealed at a meeting that while外界关注 DeepSeek's business model落地 and technical progress, the company has been making many efforts and attempts, preliminarily validating some paths.
Compared to other model companies and internet giants, DeepSeek's progress on the product side has been slow, with previous focus一直放在 model training. Regarding whether DeepSeek should develop products, Hu Jie stated that Liang Wenfeng大概 realized the importance of productization around late last year, but the team一直没招到 suitable leadership and struggled with the product路线. "The product matter requires Liang Wenfeng's own思考." Recently, public recruitment websites show DeepSeek opening many product-related positions, including Model Strategy Product Manager and Agent Data Strategy Engineer.
However, this doesn't necessarily signify a路线 change for DeepSeek. According to a May 22 media report, Liang Wenfeng stated at a meeting面向 investors that the company will continue advancing open-source AI models, aiming to achieve Artificial General Intelligence (AGI). In his表述, technological breakthrough is DeepSeek's core mission to突破 technical boundaries, not追求 profit.
Hu Jie noted that financing is merely a capital operation, and DeepSeek's daily operations and technical路线 are not necessarily bound by capital markets. An investor on the transaction list stated that for DeepSeek, he believes short-term commercialization isn't as important; achieving AGI is paramount, and China must have its own AGI.
(At the request of interviewees for anonymity, Hu Jie, Chen Duo, Li Jing, and Tang Yu are pseudonyms.)
Comments