DeepSeek reportedly initiates IPO preparations with CITIC Securities, targeting 2027 listing

Chinese AI startup DeepSeek (深度求索) is reportedly advancing its IPO preparations in parallel with a new round of financing, according to industry sources.

Funding and Valuation

The company is seeking a valuation of 500 billion yuan ($68.5 billion) in its latest funding round. Investors are expected to include existing shareholders from the first round and candidates who missed the earlier opportunity.

DeepSeek completed its first funding round in June, with total market estimates ranging from 50 billion to 51 billion yuan. Founder Liang Wenfeng is believed to have contributed approximately 20 billion yuan as the largest single investor. Other participants include:

• Tencent: Approximately 10 billion yuan
• CATL and Puquan Capital
• NetEase and JD.com

IPO Timeline

Reports from late August indicated DeepSeek is preparing for a potential IPO, with the earliest filing expected by the end of 2026 and a target listing in 2027.

Team Expansion

On September 7, 2026, Cui Tianyi, technical lead of the DeepSeek Harness team, posted recruitment notices on social media seeking approximately 150 new hires globally. Notably, the positions focus on engineering roles such as server-side development and Agent elastic computing, rather than deep learning research.

Cui explained that scaling in computing inevitably leads to exponential complexity increases. The rapid growth in data volume, machine and container counts, training and evaluation tasks, Agent environments, users, and requests has placed significant strain on existing backend systems, necessitating team expansion for upgrades, maintenance, and rewrites.

Policy Support

Analysts note that DeepSeek's IPO preparations align with favorable policy developments. On June 17, 2026, the Shanghai Stock Exchange issued guidelines supporting AI companies that have not yet achieved stable profitability in their A-share listings. DeepSeek initiated IPO preparations within three months of the policy release, demonstrating alignment with regulatory frameworks.

Commercialization Challenges

Industry observers view the IPO as a critical shift from "technology-driven" to "capital-driven" growth, signaling DeepSeek's need for substantial external funding to support computing power, R&D, and talent acquisition.

While listing does not require immediate profitability—the Shanghai Stock Exchange's June 2026 guidelines allow AI companies without significant revenue to use the STAR Market's fifth set of standards—the market will demand answers to practical questions:

• Where will revenue come from?
• Are API calls driven by genuine production needs or subsidized traffic?
• Does high user adoption translate into official revenue?

Financial Data

Reports suggest DeepSeek generated approximately 475 million yuan in revenue during the first seven months of 2026, roughly 10 times its full-year 2025 revenue. Gross margin for the period was 44.6%, with API business gross margin at 82.9%. Sources caution that these figures are not from audited financial statements and should be verified against future filings or prospectuses.

Investment bankers indicate DeepSeek may pursue full-year financial reporting rather than the industry-standard ARR (Annualized Recurring Revenue) approach for its IPO prospectus. Scaling operations and confirming revenue will be immediate priorities for Liang Wenfeng and the company.

Key Challenges Ahead

Cost Structure: DeepSeek's core advantage lies in architectural innovation that lowers inference costs. However, low costs in technical reports do not equate to high margins in financial statements. R&D experimentation, peak computing redundancy, and enterprise services will all impact the income statement.

Pricing Pressure: As price wars intensify, lower prices may expand the market but reduce per-token margins. Sustainable efficiency gains require simultaneous improvements in call volume, repurchase rates, and computing utilization.

Talent Retention: Between the second half of 2025 and early 2026, several key technical staff departed. Competition from major tech companies for top AI researchers has driven talent costs to new heights. Post-IPO equity incentives may help address retention.

Industry Implications

DeepSeek's listing could reshape valuation logic across the sector:

• Capital is expected to concentrate among leading players
• Foundation-layer companies must demonstrate both model efficiency and commercialization capability
• Application-layer firms need industry data and customer access
• Companies without technological differentiation will face pressure

Computing procurement evaluation will shift from celebrating "10,000-GPU clusters" to scrutinizing GPU utilization, unit inference costs, and real business output.

Talent valuation will also change—those who can connect models, inference systems, and customer needs on a single profit-and-loss statement will become increasingly valuable.

Regulatory and Market Risks

CITIC Securities has entered due diligence, but no formal advisory agreement has been signed. No listing registration information has been found with the Zhejiang Securities Regulatory Bureau. Given the minimum three-month辅导 process, filing within the year would be tight.

Market valuation expectations are high—pre-investment valuation in the primary market has reached 500 billion yuan, while first-seven-month revenue stands at 475 million yuan, implying a price-to-sales ratio exceeding 1,000 times. Whether secondary markets will accept this valuation remains uncertain. Speculative investment through multi-layer SPV structures has emerged, and Liang Wenfeng is personally reviewing shareholder lists to prevent complex ownership structures from impeding the IPO process.

Conclusion

DeepSeek's IPO represents a milestone for the industry—a transition from "proving we can build it" to "answering whether we can make money long-term." Its commercialization data and regulatory progress warrant continued attention.

(Source: China Business Journal)