Goldman Sachs Initiates Coverage of MiniMax: The Most Globally-Oriented Chinese Large Model!

Goldman Sachs, in its latest research report, believes that MiniMax is one of the best-positioned AI model companies in China, with significant global total addressable market (TAM) growth potential in text and programming fields.

According to ChaseWind Trading Desk, on February 23, Goldman Sachs’ latest research report initiated coverage on China's full-modal AI company MiniMax, assigning a “Neutral” rating. Based on a DCF model, it set a 12-month target price of HKD 1,018, corresponding to a company valuation of about USD 38.9 billion (RMB 282 billion).

Goldman Sachs points out that MiniMax, as one of the world’s leading AI model companies, is in a high-growth phase. Its key highlights are a highly globalized revenue structure (70% revenue from overseas) and a comprehensive multimodal product line (covering text, video, audio, music, and image). Goldman Sachs emphasizes that MiniMax occupies an excellent competitive position in the broad global TAM across text/code, multimodality, and AI agents/digital workforce fields.

Core Impact on Investors:

Valuation Elasticity Tied to Market Share: Goldman Sachs gives bullish/base/bear valuations of USD 66 billion/USD 42 billion/USD 16 billion, mainly depending on the expansion of its global AI foundational model subscription and API revenue market share (base case forecast to reach 2.5% by 2030, bull case at 5.0%).Commercialization and Profitability Path: The company’s revenue is expected to surge from USD 75 million in 2025 to USD 980 million in 2027. Although it is still making losses (expected adjusted net losses for 2025-2027 are USD 316/413/388 million), its gross margin is expected to rise from 21% in 2024 to 38% in 2027. The company is well-funded (over USD 1 billion in cash reserves) and sufficient to sustain operations until the projected 2029 break-even point.Intensive Technology Catalysts: The release of the M2.5 model has significantly narrowed the gap with US SOTA (state-of-the-art) models, at just one-tenth of the cost. Subsequent launches of the Hailuo 3.0 video model and mass adoption of AI agents at only USD 1 per hour will serve as important valuation catalysts.

Multimodal Advantages & Globalized Commercial Map

Full-modal presence and up to 70% overseas revenue ratio build a core moat.

Since its inception, MiniMax has insisted on a multimodal fusion approach and currently owns leading models such as M2.5 (base text model), Hailuo 2.3 (video), Speech 2.6 (audio), Music-2.5, and Image-01. This native multimodal fusion technology (such as text models supporting video generation) gives it significant advantages in instruction following, semantic understanding, and physical interaction.

On commercialization, MiniMax targeted the global market from day one. Currently, 70% of its revenue comes from overseas. Its AI-native products have served over 212 million individual users in more than 200 countries and regions worldwide, and its base models serve over 130,000 enterprises and developers. The company has established diversified monetization channels such as subscriptions, token billing, online marketing, and API calls.

Extreme Cost Efficiency & Foundational Architecture Innovation

With MoE architecture and linear attention mechanism, it approaches world-class levels at extremely low cost.

Goldman Sachs estimates MiniMax has spent only about USD 420 million (estimated from 2022-2025) in training costs and has already joined the ranks of top global AI companies. This is thanks to its early adoption of a mixture-of-experts (MoE) architecture and linear attention mechanism, which dramatically reduce consumption of computing resources and inference cost.

This cost advantage directly translates to competitive product pricing. For example, its API pricing is only about 8% of that of overseas leading models (such as Claude 4.5 Sonnet), making its To-B API business able to maintain healthy gross margins despite large price cuts (up to 69% gross margin for open platforms in the first 9 months of 2025). Additionally, its flattened organizational structure (about 300 R&D staff) ensures high innovation efficiency.

Agent Explosion and “$1 Digital Workforce”

M2.5 model performance approaches Claude Opus; agent cost at USD 1 per hour may ignite application boom.

The recently launched M2.5 model’s performance in programming and office scenarios has further improved. Its SWE-bench Verified coding score reached 80.2%, very close to Anthropic’s strongest model Opus 4.6 (80.8%), with only one-tenth the cost.

With such high cost-performance, MiniMax Agent applications demonstrate great disruptive potential. Goldman Sachs estimates MiniMax agent operating costs at just USD 1 per hour (at 100 tokens per second). This means that a USD 10,000 budget could hire 4 “digital employees” to work around the clock, year-round. This compelling economics is expected to drive AI adoption in enterprise scenarios like HR screening, code debugging, etc.

Financial Forecast & Valuation Logic

Revenue aiming at USD 980 million in 2027, multimodal API driving sustained gross margin increases.

Goldman Sachs forecasts MiniMax’s revenue to grow from USD 75 million in 2025 to USD 980 million in 2027. The core drivers of revenue growth include:

Hailuo AI (Video): Revenue projected to reach USD 311 million in 2027.Open API Platform: With pricing competitiveness and multimodal capabilities, revenue projected at USD 379 million in 2027.Talkie/Xingye (Consumer Entertainment): Revenue projected at USD 184 million in 2027.

Despite high costs for R&D and inference leading to short-term losses (adjusted net profit forecast at USD -388 million in 2027), gross margin is expected to leap from 21% in 2024 to 38% in 2027, mainly due to the rising proportion of high-margin multimodal API revenue (75% of total API revenue).

For valuation, under the base scenario (HKD 1,018/USD 38.9 billion), it’s assumed the company secures 2.5% of the global foundational model subscription and API market by 2030; in the bullish scenario (HKD 1,600/USD 66 billion), market share accelerates to 5.0%, representing a 44x price-to-sales ratio of ARR (annual recurring revenue) in Q4 2027; in the bearish scenario (HKD 380/USD 16 billion), intensified competition leaves market share at only around 1.2%.

~~~~~~~~~~~~~~~~~~~~~~~~

The above exciting content is from ChaseWind Trading Desk.

For more detailed interpretation, including real-time analysis and frontline research content, please join [ChaseWind Trading Desk▪Annual Membership]

Risk Warning and DisclaimerThe market has risks and investment needs caution. This article does not constitute individual investment advice and has not taken into account individual users’ special investment goals, financial situations, or needs. Users should consider whether any opinions, viewpoints, or conclusions in this article suit their particular situation. Invest at your own risk.