Zhipu VS MiniMax: The Capital Logic and Investment Value of the Battle for the 'First Stock' of Large Models

Zhipu and MiniMax (Xiyu Technology) have successively advanced their IPO processes. The battle for the "first stock" in the field of large models is not only a capital competition between two unicorn companies but will also set valuation benchmarks for the AI sector in the A-share and Hong Kong markets. The listing of these two companies fills the gap in the capital market for pure large-model targets, and the differences in their business models, technological barriers, and commercialization capabilities will also determine investors' strategic logic in the AI race.

I. Dual Titans: Core Differences in Technical Paths and Business Layouts

Although both Zhipu and MiniMax belong to the large-model race, their technical routes and commercialization focuses are 截然不同， which becomes the key to distinguishing their investment value:

1. Zhipu: The "Technical Faction" Backed by Research
   Zhipu was incubated by a team from Tsinghua University, with core technologies derived from top-tier research achievements in the field of natural language processing. It focuses on the general large model Zhipu Qingyan while also venturing into To B businesses such as AI computing power leasing and industry solutions (finance, education). Its advantage lies in its deep technical foundation and the continuous technological iteration capability formed through university-industry collaboration. However, its commercialization pace is relatively slow, and the user scale and monetization efficiency of its To C products have yet to form a significant advantage.
2. MiniMax: The "Practical Faction" Focused on Scenarios
   MiniMax centers on conversational large models, launching To C products like Doubao (in collaboration with ByteDance) and MiniMax-Chat, while also providing customized large-model services for industries such as internet and entertainment. Its characteristic is more flexible commercialization, with rapid user growth on the To C side and stable traffic and income from collaborations with major companies. However, its underlying technical R&D 积淀 is relatively weak, and it lags behind Zhipu in terms of parameter scale and multimodal capabilities of large models.

II. The Capital Valuation Logic of the "First Stock" in Large Models

The capital market's valuation of large-model companies is not solely based on technical strength but rather a comprehensive consideration of technological barriers + commercialization potential + industry scarcity:

1. Industry Scarcity: The Core Premium Point of Valuation
   Currently, AI targets in the A-share market are mostly hardware and application-layer companies, with pure large-model R&D companies in a 空白 state. The listing of Zhipu and MiniMax will gain a "lone 苗" valuation premium. Referring to the valuation logic of overseas OpenAI-related concept stocks, their market capitalization is expected to approach that of AI sector leaders.
2. Commercialization Capability: The Watershed of Valuation
   Zhipu: The income stability of its To B business is strong, but the order 落地 cycle is long, with weak short-term profit expectations, making its valuation more 偏向 "technology premium."
   MiniMax: The monetization efficiency of its To C products (advertising, value-added services) is high, and collaborations with major companies can quickly 兑现 income, making its valuation 更容易获得 "performance support."
3. Policy and Computing Power: Potential Risks to Valuation
   The large-model industry is significantly affected by computing resources and data compliance policies. If future computing power bottlenecks or stricter data regulations emerge, they will directly suppress the valuation space of these two companies.

III. Investors' Layout Strategy: Grasp the Main Line, Discern Differences

1. Short-Term: Focus on the Expectation Gap in IPO Pricing
   If the IPO pricing of Zhipu or MiniMax is lower than market expectations, investors can participate in new subscriptions or short-term 波段 operations to earn a "scarcity premium." If the pricing is too high, be wary of post-listing valuation 回归 risks.
2. Medium- to Long-Term: Focus on the Certainty of Commercialization 落地
   Layout for MiniMax: More suitable for investors who prefer "fast performance 兑现，"关注其 To C products' user growth and advertising income data.
   Layout for Zhipu: More suitable for long-term value investors, tracking the 落地 of its To B industry solution orders and the profitability of its computing power leasing business.
3. Industry Chain 联动：挖掘细分赛道 Opportunities
   Regardless of who becomes the "first stock in large models," it will drive valuation 修复 in the AI industry chain. Investors can simultaneously 布局细分赛道 such as computing hardware (GPUs, servers), data annotation, and AI applications to share industry 红利。

The IPO competition between Zhipu and MiniMax is essentially a capital showdown between the "technical faction" and the "practical faction" in the large-model industry. For investors, there is no need to 纠结 over "who goes public first." Instead, focus on the commercialization capabilities and technological barriers of both, and in the long-term development of the AI race, find investment targets that combine growth and certainty.$C3.AI(AI.US)