Inside MiniMax’s Meteoric Rise: 1 Million Enterprise Clients and 300 Million Users in Six Months

Enterprise and Developer Clients: From 200,000 to 1 Million

The jump from approximately 200,000 to over 1 million enterprise and developer clients represents a fivefold increase . This metric is particularly significant because it reflects adoption among businesses and developers who integrate MiniMax's AI models into their own products and workflows—a stickier, higher-value customer segment than casual consumer users.

The growth was disclosed by MiniMax co-founder and president Yun Yeyi during his appearance at the UBS conference, underscoring the company's aggressive push beyond China's borders .

Global User Base: 300 Million and Counting

MiniMax now claims approximately 300 million users worldwide . For context, this places the company among the largest AI platforms globally. While the figure includes both free and paying users, the sheer scale signals that MiniMax's consumer-facing AI products have achieved mass adoption, particularly in international markets where the company has been investing heavily.

ARR Doubling: A $150 Million Milestone

The most striking financial detail is the ARR trajectory. MiniMax reported that its annual recurring revenue had more than doubled in just two months . The company had previously disclosed at a March 2026 earnings briefing that ARR had exceeded US$150 million as of February 2026, itself a significant jump from the $100 million level it reached earlier in the year . Put another way, MiniMax added roughly $50 million in annualized revenue in approximately eight weeks.

This acceleration is part of a longer trend. MiniMax's total revenue for the 12 months ending December 31, 2025, climbed 159% to $79 million, with about 73% of that coming from overseas markets . Most of the revenue originated from AI-native products rather than traditional enterprise software sales, signaling that MiniMax's AI-first strategy is monetizing effectively .

The Red Ink Beneath the Growth

For all its top-line momentum, MiniMax is not yet profitable. The company is grappling with widening financial losses, a disclosure that accompanied the SCMP report . The tension between rapid scaling and mounting costs is familiar across the AI industry, but it is particularly acute for Chinese AI firms operating in a market where competition is fierce and model training costs remain enormous.

MiniMax listed on the Hong Kong Stock Exchange on January 9, 2026, at an IPO price of HKD 165, and its stock has since gained roughly fivefold . As of late May 2026, its market capitalization exceeded HKD 250 billion (approximately RMB 223.5 billion), nearly a fourfold increase in the four months since listing . Despite the red ink, investors are betting heavily that the company's growth trajectory will eventually translate into sustainable profits.

The Bigger Picture: China's AI Tiger Race

MiniMax is not alone in posting eye-catching numbers. Other members of China's "AI Tigers" cohort have been racing to raise capital and scale users. Moonshot AI, the maker of the Kimi chatbot, saw its valuation jump from $4.3 billion to $20 billion in six months and reported its own ARR doubling from $100 million to over $200 million in just two months .

But MiniMax's disclosure stands out for the sheer breadth of its metrics: enterprise clients, consumer users, and revenue all accelerating simultaneously. The company's ability to scale all three at once, while maintaining an international-first revenue mix, suggests it has found product-market fit beyond China's borders faster than many of its peers.

The question that hangs over all of these firms, however, is how long investors will tolerate the widening losses. By late May 2026, according to one market observer, investors were "frantically" snapping up shares of Chinese AI companies despite their lack of profits . Whether that enthusiasm endures will depend on whether companies like MiniMax can eventually convert their explosive user and client growth into sustained profitability—something no major AI company has yet convincingly demonstrated.