A familiar prospectus has been posted on the HKEX website.
On May 28, following the invalidation of its first submission in October last year, Intelligent Driving Solutions Provider Suzhou Tantong Vision Electronic Technology Co., Ltd. (Tantong Vision) launched another bid for the HKEX Main Board, with HSBC and Huatai International as joint lead managers.
In 2026, as the intelligent driving track shifts from 'storytelling' to 'mass production', Tantong Vision's second submission is not only a capital probe, but also a comprehensive review of the survival status of Chinese intelligent driving suppliers.
This company, regarded as a representative of the 'Computing Efficiency School', is connected to a prestigious industrial capital camp ranging from ZF to SAIC and BAIC, while simultaneously facing the realistic dilemma of tight cash flow and a significant decline overseas. In this IPO game, brilliance and pain coexist.
01
Who is 'Tantong Vision'?
Wang Xi, the founder of Tantong Vision, is a typical 'returnee' technical expert. He graduated from Beihang University and later pursued a PhD in Computer Science at the University of Reading in the UK.
Before deciding to start a business, Wang Xi worked as an algorithm engineer and technical lead at automotive parts supplier TRW Automotive and ZF, deeply participating in the development of early ADAS systems.
In 2016, Wang Xi seized the domestic automotive intelligence hotspot and returned to China to establish Tantong Vision in Suzhou, positioning it as a 'software algorithm-driven intelligent driving' local solutions provider.
The company's name 'Tantong' implies 'Eye of the Sky', aiming to build a visual hub for vehicles to perceive everything. Starting with visual perception algorithms, it gradually expanded to integrated driving and parking domain controllers, L4 level autonomous driving systems, and other overall solutions combining hardware and software.
Tantong Vision Financing Situation. Source: Qichacha
Shortly after establishment, Tantong Vision received angel round financing from Delian Capital and Shengshi Investment. In the following ten years, Tantong Vision completed more than 10 financing rounds in total, with a financing amount close to 1 billion yuan.
From the equity structure disclosed in the prospectus, Tantong Vision built a deeply bound 'Industry + Capital' ecosystem.
On the one hand, industry partners provide backing. Global automotive parts giant ZF is not only its Series C lead investor but also its strategic partner, holding 6.93% of Tantong Vision shares, ranking as the fourth largest shareholder; domestically, SAIC Group holds shares through SAIC North America Investment, and BAIC Group is involved through BAIC Investment, with Horizon Robotics and SenseTime also being strategic investors.
On the other hand, local state-owned assets provide escort. State-owned background funds such as Tangshan Robot Fund and Wujin Financial Holdings entered in Series D and D+ rounds, providing about 523 million yuan in capital support.
Equity Structure as of the Last Practicable Date
As of now, Wang Xi controls approximately 40.84% of the company's equity through direct shareholding and employee shareholding platforms, maintaining control of the company.
02
Walking with Two Legs
Tantong Vision adopted a 'Dual-Track Parallel' strategy in its business layout.
In the L2-L2+ Level Advanced Driver Assistance field, Tantong Vision's choice was very pragmatic. It did not blindly chase the 'Arms Race' of computing stacking, but walked a high cost-performance route.
As a typical Vision-based Intelligent Driving Supplier, its L2 mass production solutions focus on visual perception, integrating millimeter-wave radar and ultrasonic radar, capable of achieving advanced functions on lower computing power platforms. For example, solutions based on the Horizon J6B chip (about 20 TOPS) can support integrated driving and parking, Highway NOA.
This approach hits the pain point of the 100,000 to 200,000 yuan mainstream model segment being cost-sensitive. According to data from Frost & Sullivan, based on 2024 installed volume, Tantong Vision is China's second-largest software-centric L2-L2+ solutions provider providing both driving and parking solutions, with a market share of 14.3%.
2024 China Software-Centric Suppliers with Integrated Driving and Parking Capabilities Landscape
As of the Last Practicable Date, Tantong Vision received L2-L2+ solution designation letters for 198 models from 23 car brands, and achieved mass production for 105 models from 6 car brands; among the 198 models with designation letters, 87 cover overseas markets, of which 59 have achieved mass production.
However, it is worth noting that the L2-L2+ market is undergoing intense 'Red Oceanization'.
On the one hand, local Tier 1s such as Jingwei Hengrun and Fortec are accelerating their chase; on the other hand, some head car companies are starting to switch low-level intelligent driving solutions from outsourcing to internal integration.
Whether Tantong Vision can maintain its leading position in the 'Cost-Effective Solutions' field depends on whether it can continuously maintain the optimization capability of algorithms on low computing platforms, which requires continuous investment in R&D.
In the high-level L4 Level Autonomous Driving field, Tantong Vision plays more of a 'Pioneer' role. This is also the fastest-growing sector for it in the last two years.
As early as 2019, it participated in the Shanghai Yangshan Port 5G Intelligent Heavy Truck project. Currently, its L4 solutions cover Robobus, Robotaxi, and Robotruck. Among them, Robobus is its most representative product line, already in regular trial operation on public roads in cities such as Suzhou and Tianjin.
In 2025, Tantong Vision generated revenue of 375 million yuan from L4 Level solutions, accounting for more than 68% of the company's total revenue, with most revenue coming from L4 Level software solutions.
However, on the other side of the coin are the twists and turns of commercialization. Although L4 business revenue surged, its delivery form is currently mainly 'Hardware-Software Integrated Solutions', a model that is essentially close to 'Project-based Delivery' or 'Small Fleet Deployment', which has significant differences from the high gross margin, mass replication logic of 'Pure Software Licensing + White-box Delivery' in L2 business.
This directly led to large fluctuations in the gross margin of L4 business: in some projects with a high proportion of self-developed hardware, the gross margin once dropped to 15%.
As of the submission date, although the company holds over 1 billion yuan in L4 Letters of Intent Orders covering 2,500 vehicles, these orders are expected to be delivered successively within the next three to five years, with limited improvement in cash flow in the short term.
In addition, Tantong Vision also has some receivables from engineering services, mainly involving road testing, data collection support and data annotation services, as well as the company's proprietary toolchain.
03
Financial Double-Sided Mirror
The financial section of the prospectus reveals the most realistic 'Dark Side' of the intelligent driving industry: the extreme struggle between scale and losses, and the tense game between book cash and operational burn.
Revenue high growth, but structure fluctuates violently.
Financial data shows that the company's revenue grew explosively, from 172 million yuan in 2022 to 483 million yuan in 2024, with a CAGR of 67.7%. Full-year 2025 revenue grew further to 550 million yuan.
But the change in revenue structure is obvious. In 2023, the company relied on L2-L2+ business, accounting for 90.2%; by 2024, the L4 business share rose to 50.2%; in 2025, the L4 business share further rose to 68%. This 'Cliff-like' structural switch, while proving its L4 technology found landing scenarios, also led the market to question whether its L2 business has hit the ceiling.
The divergence of gross margin and net profit is the biggest challenge Tantong Vision faces.
From gross margin, the overall gross margin hovers around 30%, which is a medium level in the technology-intensive intelligent driving industry. But breaking it down, the gross margin of L2-L2+ business usually maintains above 40%, the pure software licensing model, while the gross margin of L4 business is significantly pulled down by the increased hardware proportion in 'Hardware-Software Integrated' delivery.
From net profit, although the surface loss is large, 463 million losses in 2024 and about 200 million losses in 2025, this includes a large amount of 'Paper Losses' caused by changes in the fair value of preferred shares. The Adjusted Net Profit after excluding this factor better reflects the company's real operating status: it narrowed to -4.38 million yuan in 2024, but in 2025 it did not turn positive as the market expected, but recorded about -10.86 million yuan, with losses expanding compared to 2024.
Behind this is an inescapable key premise: the adjusted 'loss reduction' even 'close to break-even', was achieved on the basis of the company continuously compressing R&D investment. R&D expenses dropped from 117 million yuan in 2024 to 92.31 million yuan in 2025, the R&D expense ratio dropped further from 24.3% in 2024 to 16.8%, while in 2022 this figure was as high as 108.7%. For a technology company, whether the 'decline' in R&D intensity will affect future technical moats is a potential risk point.
Cash flow continues to be urgent, the most worrying signal.
According to the latest prospectus, as of December 31, 2025, the company's cash and cash equivalents on the books were 235 million yuan, a net decrease of 139 million yuan compared to 374 million yuan on June 30, 2025, with a faster cash burn rate.
More worth alerting is the trend of operating cash flow turning from positive to negative and the gap continuously expanding. In 2023, the company's operating cash flow net amount was a positive inflow of 115 million yuan, but in 2024 it quickly turned to a net outflow of 189 million yuan, and in 2025 it further deteriorated to a net outflow of 293 million yuan.
At the same time, accounts receivable turnover deteriorated sharply. The company's trade receivables rose from 89 million yuan in 2023 to 548 million yuan in 2025, a growth of more than five times in three years, while the revenue increase in the same period was only about 2.7 times.
More worrying is that the accounts receivable turnover days jumped from 191 days in 2023 and 166 days in 2024 to 300 days in 2025, meaning that the company needs an average of nearly a year from completing delivery to recovering funds. This is equivalent to providing long-term interest-free funding for customers, further aggravating liquidity pressure when funds were already tight.
In addition, although Tantong Vision is the first Chinese intelligent driving software provider to go overseas, its overseas business experienced a significant decline in 2025. In 2023, Tantong Vision's overseas revenue was 127 million yuan, accounting for 62.2% of total revenue; by 2025, overseas revenue dropped to 11 million yuan, accounting for only 2.0%.
This adds a touch of uncertainty to its globalization story against the backdrop of complex global geopolitics and stricter data regulation on intelligent cars in some countries.
04
Conclusion
Tantong Vision's second attempt is a microcosm of the intelligent driving industry entering the 'Elimination Round' stage.
On the positive side, it hit the dual rhythm of L2 cost-effectiveness and L4 scenario landing, and the adjusted net profit is close to break-even in a specific measure, which provided investors with a 'storyline with highlights to tell'.
But from a risk perspective, the situation is far more severe than the first submission. The rapid drift of business focus, the profitability difficulties in the early stage of L4 business commercialization, and the passive contraction of R&D investment, these problems that already existed have not been alleviated.
Tantong Vision L4 Level Intelligent Bus
What truly gives this IPO a 'Survival' color is the substantial deterioration of cash flow data: 235 million yuan in book cash, facing an annual operating cash net outflow of nearly 300 million yuan, the safety margin is less than one year. Superimposed on 300 days of accounts receivable turnover, it means that for every order delivered, the company has to advance nearly a year's cost of capital.
In other words, Tantong Vision is in a dangerous financial window period: The money on hand is only enough to maintain normal operations for less than a year, while the mass delivery and repayment of L4 business requires more time. At a time when the intelligent driving capital heat has subsided and primary market financing has tightened marginally, the company has no more room to wait.
Only six months after the first submission failed, it again challenges the HK stock market. For Tantong Vision, rather than a strategic choice, it is an inevitable move forced by cash.
© 2024 Bitauto.com. All Rights Reserved.
Bitauto (www.yiche.com and www.bitauto.com) is one of the most visited automotive websites in China. We provide detailed info on Chinese car brands available in your area. Please note that the information provided on this website is based on publicly available data and is for reference only. It may not be fully updated or accurate at all times.
