MiniMax Knocks on the Door of Hong Kong Stock Exchange with Billion-Dollar Valuation

Key Takeaways

• MiniMax, a prominent AI startup, is rapidly progressing towards an IPO on the Hong Kong Stock Exchange, potentially setting a record for the swiftest transition from inception to IPO in the AI industry.
• The core of MiniMax’s success lies in its dual approach: integrating large AI models with native applications, which has attracted over two billion users worldwide.
• Despite high research and development investments, MiniMax showcases financial efficiency with significant global penetration, particularly securing over 70% of its revenue from international markets.
• The company’s youthful leadership, with an average age of 32 among its directors, reflects its innovative edge and adaptability in the fast-evolving AI landscape.

WEEX Crypto News, 2025-12-22 16:06:41

The pace at which AI startups are advancing towards Initial Public Offerings (IPOs) seems to be accelerating at an unprecedented rate. This week, MiniMax, a notable AI startup based in Shanghai, reached a pivotal moment in its journey to become a public company by publishing its Post Hearing Information Pack (PHIP) version of the prospectus on the Hong Kong Stock Exchange. Having been founded in December 2021 by former executives from SenseTime, including Junjie Yan, MiniMax has, in its short history, carved out a reputation as a pioneer in developing comprehensive artificial intelligence (AI) solutions that integrate text, voice, and visual modalities—typically referred to as Artificial General Intelligence (AGI).

Reaching for Records with Remarkable Efficiency

MiniMax’s potential IPO is not just noteworthy for the prospect of capital-raising, but also for the company’s potential to become the fastest AI entity to transition from inception to IPO. When MiniMax was established, it staked its claim on the competitive landscape with its flagship products—MiniMax M1, M2 foundational models, and AI-native solutions like MiniMax Voice, Hailuo AI, and Xingye. The company’s ability to swiftly scale and innovate reflects its commitment to maintaining a leading edge in the burgeoning AI sector.

MiniMax has disclosed comprehensive details of its business framework, revenue streams, and strategic positioning globally through its PHIP. This transparency highlights MiniMax’s strong foothold in the global AI arena, where it boasts over two billion users and a commanding presence in the AI developmental frontier.

Global Reach and Dynamic Growth

MiniMax has positioned itself as a frontrunner among AI companies by implementing a dual-pronged strategy focused on “big model + AI native applications.” This approach is evident in the success of its platform, which has emerged as a significant player in the global enterprise and developer space, facilitating integration across industries such as smart devices, healthcare, financial services, and more. The company reported over 2.12 billion individual users and more than 100,000 enterprises and developers across 200 countries as of September 2025. This vast user base signifies not only the scalability of MiniMax’s models but also their versatile application.

International markets contribute approximately 70% of MiniMax’s revenue, underscoring the company’s effective global expansion and strategic market engagement. Financial metrics reveal extraordinary growth, with the company achieving $53.437 million in revenue during the first nine months of 2025—an increase of 174.7% compared to the prior year. Earnings are primarily driven by two streams: subscriptions and in-app purchases of AI native products, alongside enterprise services through their open platform.

A key metric illustrating MiniMax’s operational excellence is its accounts receivable turnover, which stands at 38 days. This figure is significantly lower than the average of 60 to 90 days typical in AI or SaaS industries, indicating superior cash flow management and business efficiency.

Expanding Horizons Through Strategic Alliances

In the consumer segment, MiniMax develops and delivers products such as Hailuo AI, Xingye (Talkie for AI social networking), and MiniMax Voice to consumers worldwide. The consumer segment saw a revenue spike of 181% year-over-year, with its paid user count surging fifteenfold within just two years.

In terms of business solutions, MiniMax empowers various industries through its open platform, making its advanced model capabilities available via Application Programming Interfaces (APIs). The platform manages over a trillion token requests daily and has seen a 160% growth in business-to-business (B2B) income, boasting a gross margin of 69.4%. These figures illustrate a robust potential for profitability within the high-demand sector of AI solutions.

Globally, MiniMax has forged strategic partnerships with leading technology firms, enhancing their technological ecosystem. For instance, its models operate on the AI platforms of major cloud services providers such as Amazon, Google, and Microsoft and are incorporated in consumer products of companies like LinkedIn and Monks, a digital creativity hub. Domestically, MiniMax collaborates with giants like Alibaba, Tencent, ByteDance, Xiaomi, and Kingsoft, integrating its capabilities into core products and even providing human-like interaction features for humanoid robots.

Commitment to All-Encompassing AI Technologies

The trajectory of MiniMax is grounded in an effective and forward-thinking strategy: the development of an all-encompassing AI technology matrix. MiniMax is among the few startups that, from their inception, dedicated themselves to creating multimodal AI models. The company’s intense focus on continuous technological iteration has led to noteworthy advancements in areas such as speech, video, and text—capturing the attention of the tech community globally.

Since 2023, MiniMax has introduced groundbreaking models like Speech 01 and its successor, Speech 02, utilizing the Transformer architecture. By 2025, MiniMax’s voice models supported users in generating over 2.2 billion hours of speech. Video model innovation followed, with the release of Video 01 and Hailuo AI in 2024, culminating in an update—Hailuo 02, which secured a top-tier ranking in video generation competitions.

Another milestone was MiniMax’s October 2025 launch of the open-source text model, MiniMax M2, ascending to the top five in the Artificial Analysis rankings, marking the first Chinese open-source model to achieve such global distinction. This showcases a relentless progression in model development across text, video, voice, and even music.

The standout text model MiniMax-M2, launched on October 27, impressed with reasoning speeds nearly double those of its closest competitor, Claude Sonnet 4.5, at just an eighth of the price. This model’s integration into Amazon Bedrock, AWS’s fully managed generative AI cloud service, cements its status as a landmark domestic AI model.

MiniMax M2 employs a method known as “Interleaved Thinking,” which has sparked conversation in AI research communities abroad. This technique enables the model to engage in cycles of “thinking-action-reflection,” allowing for continuous context comprehension and real-time strategy adjustment. The result is an agent with enhanced planning, stable execution capabilities, and reliable self-correction, offering significant advantages in handling complex tasks.

Pursuit of Efficiency Amidst Intensive R&D Investments

As with many AI firms, MiniMax remains in a phase of high R&D expenditure and rapid prototype iteration, yet profitability has not been accomplished. Through the first nine months of 2025, net losses amounted to $186 million, modestly up by 8.6% after adjustments. These losses stem mainly from R&D expenses and related financial obligations.

Despite the hefty R&D investment which amounted to $180 million (337.4% of total income), the growth in revenue far outpaced expenditure, highlighting effective operational leverage. This underscores the company’s strategic foresight in prioritizing essential model iteration and foundational infrastructure.

MiniMax’s organizational architecture emphasizes productivity and streamlined operations. With approximately 385 employees, 300 are involved in R&D—reflecting the company’s commitment to innovation. Predominantly composed of “post-95s” and “post-00s,” this youthful workforce, with members previously part of elite global tech firms such as Microsoft, Google, Meta, and Alibaba, aligns well with the current AI development paradigm. This paradigm champions new tools, automated processes, and AI-native methodologies, which naturally enhances productivity per capita.

The company’s management style, rooted in a “minimalist” approach, limits hierarchical layers below the CEO to just three, facilitating rapid project execution and maintaining a high cost-effectiveness ratio amidst intense computational and financial competition. This strategic focus within MiniMax on leveraging AI capabilities has realized an extraordinary development journey—from leading in a single modality to excelling across multiple, forming a cohesive and interoperable technological framework.

A Bold Venture Towards Advanced Generative Intelligence

MiniMax’s bid for an IPO is not merely an indication of yet another AI unicorn pursuing capitalization. It embodies a narrative about AGI, woven with youth and ambition. Its team, led by directors averaging just 32 years old, leverages AI-native strategies to build innovative enterprises from the ground up. Driven by their recognition of the transformative potential of seamless modality integration and a focus on global markets, MiniMax’s bold stride forward embraces both the challenges and opportunities inherent in pioneering a new sector.

With the momentum demonstrated by its documented achievements, MiniMax continues to progress towards its long-term vision of democratized intelligence. This vision of delivering “intelligence with everyone” will likely persist as a driving force in its ongoing development. As MiniMax forges ahead to redefine boundaries within the AI industry, it offers a remarkable case study for future iterations of technology-driven business ventures.

FAQs

What are MiniMax’s main AI products?

MiniMax is renowned for its foundational AI models like MiniMax M1 and M2, in addition to applications such as MiniMax Voice, Hailuo AI for video generation, and Xingye for social networking.

How does MiniMax generate the majority of its revenue?

MiniMax mainly derives its revenue from subscriptions and purchases of its AI-native products, and services offered on its open platform for enterprises.

Why is MiniMax considered a leader in the AI domain?

MiniMax’s integration of multimodal AI models and rapid technology iteration has placed it at the forefront of AI development. Its emphasis on deploying comprehensive AGI solutions across global markets underscores its leadership position.

What distinguishes MiniMax’s business approach from other AI companies?

MiniMax employs a unique strategy by combining large AI models with native applications, focusing on global expansion, and achieving significant financial efficiency through high operational turnover and strategic partnerships.

What role does youth play in MiniMax’s strategic vision?

With a leadership team of “post-95s,” MiniMax embodies a youthful, innovative approach towards AGI, using cutting-edge AI-native methodologies to navigate and thrive in the ever-evolving AI landscape.