MiniMax’s Enterprise Revenue Surges as Model Companies Chase Sustainable Scale

NextFin News — MiniMax Group posted revenue of about $117 million for the six months ended June 30, up 283 percent from a year earlier and already higher than the company’s entire 2025 total. Gross profit rose nearly fivefold to $20.8 million. The net loss narrowed 11 percent to $358 million. Adjusted net loss stood at $293 million, and the adjusted loss rate fell from roughly 456 percent to about 251 percent.

The mix of that revenue matters more than the headline growth. Open-platform and other enterprise AI services generated $73.9 million, up more than sevenfold, and accounted for 63 percent of total revenue, up from 30 percent a year earlier. Revenue from AI-native products doubled to $42.6 million. Overseas markets contributed 61 percent of sales. Research and development spending rose 139 percent to $297 million—still several times revenue—while sales and distribution costs fell nearly 18 percent as the company relied more on organic adoption.

Founder and chief executive Yan Junjie framed the strategy around a simple tension: intelligence can keep improving, but energy and compute are finite. Token consumption in July, he said, reached twenty times the January level. The company’s aim remains lowering the cost of delivering higher intelligence to more users rather than maximizing model size alone.

Product updates during and after the period point in the same direction. The M3 model kept earlier pricing while improving capability and drawing more enterprise and developer workloads. Usage is shifting from single-turn human queries toward multi-step agent workflows that generate chains of model calls, tool use and sub-tasks. That pattern is driving token volume faster than growth in users or messages. After the reporting period the company released an open-source video model, H3, that quickly attracted derivative models and high download volumes.

MiniMax said its models and products now serve more than two million enterprise customers and developers—roughly ten times the figure at the end of 2025—and more than 300 million individual users across more than 230 countries and regions. Shares closed little changed on the results day at about HK$303, giving a market value near HK$106 billion.

From capability contests to serving economics

MiniMax’s numbers sit inside a broader industry transition. For several years Chinese and global model developers competed mainly on benchmark scores, parameter counts and headline capabilities. That phase produced rapid technical progress and heavy capital consumption. As models have become good enough for many practical tasks, attention has moved to a different set of variables: cost per token, utilization of serving infrastructure, the stickiness of enterprise workflows, and the ability to monetize usage driven increasingly by agents rather than individual chat sessions.

Enterprise and API revenue is becoming the clearest proof of commercial relevance. When businesses embed models in customer service, content pipelines, internal tools or multi-agent systems, consumption becomes recurring and measurable. Consumer applications can still build brand and data advantages, yet they have proven harder to convert into high-margin, predictable revenue at scale. MiniMax’s swing toward a majority enterprise mix mirrors a pattern visible among several model companies that publish segmented figures: the fastest growth and the most defensible pricing power appear where models are treated as infrastructure rather than destinations.

Overseas revenue shares above 50 percent are also becoming more common among Chinese model providers that offer competitive open weights or aggressive API pricing. Global developers and startups remain price-sensitive and willing to switch among providers that deliver acceptable quality at lower cost. That dynamic rewards continuous efficiency gains in training and inference. It also exposes strategies that rely too heavily on a single domestic market.

Losses remain structural for most pure model companies. Training runs, talent and serving capacity still require heavy outlays relative to current revenue. The more encouraging signal in recent results is not the absence of losses but the relationship between revenue growth and cost growth. When revenue expands faster than research spending and sales costs are held in check, the path to narrower losses becomes visible even if absolute break-even remains distant. Gross-margin expansion, as MiniMax reported, is an early sign that utilization and pricing are beginning to improve.

What comes next

Three pressures will shape the next phase. First, agent-driven workloads raise token intensity per user task. Providers that cannot keep unit costs falling will see margins compressed even as volume rises. Second, open-source releases speed adoption and ecosystem growth but also intensify price competition and reduce differentiation on raw capability alone. Third, enterprise customers are becoming more sophisticated about total cost of ownership, latency, reliability and data handling. Procurement conversations are moving away from pure benchmark tables.

MiniMax’s first-half results show one company navigating that shift: rapid revenue growth, a decisive move toward enterprise and API income, rising overseas contribution, improving gross profit, and continued heavy investment tempered by some operating leverage. Similar patterns are appearing, to varying degrees, across the group of model companies that have chosen public markets or detailed disclosure. The industry is no longer judged solely by who trains the most impressive model. It is increasingly judged by who can deliver usable intelligence at a cost and scale that enterprises will pay for repeatedly. That test is only beginning.