MiniMax上半年减亏增收股价上涨   2026-08-27

 

MiniMax’s shares jumped after the Chinese artificial intelligence startup shrank its loss and increased revenue in the first half of this year after shifting focus to enterprise clients.

MiniMax’s net loss narrowed 11 percent to USD358 million in the six months ended June 30 from a year earlier, the Shanghai-based company said in a financial report released on August 26. Revenue surged 283 percent to about USD120 million. In comparison, its income was USD79 million for the whole of last year.

Income from open platform and other AI enterprise services surged more than seven times to USD73.9 million, becoming MiniMax’s top revenue stream, with the growth primarily driven by a growing number of paid and enterprise clients, higher application programming interface call volumes, and rapid adoption of TokenPlan, the company said.

The annual recurring revenue topped USD800 million as of this month, Yan Junjie, founder and chief executive of MiniMax, said at an earnings conference call. While ARR is a projected annual revenue metric based on existing business volume and does not equate to recognized income, the figure signals accelerating commercialization.

Business-to-business accounted for about 80 percent of ARR in the first half of this year, with consumer-facing operations making up the rest, which is a stark reversal from the same period a year ago, when B2B contributed around 30 percent and B2C the rest, Yan pointed out. Within just one year, MiniMax has shifted its revenue mix from consumer-led to dominated by enterprise and developer offerings, he noted.

Overseas income topped USD70.8 million, accounting for 61 percent of the total and remaining a core revenue pillar for MiniMax.

However, the rapid top-line expansion did not erase all losses. For AI model companies in ongoing cycles of model training, product rollout and infrastructure buildout, computing power costs represent a major burden.

MiniMax dynamically allocates compute resources based on model maturity, business needs and marginal returns, Yan said, adding that text models are the highest priority for compute investment and receive around four times more training resources than video models.

In addition, MiniMax is diversifying its computing capacity supply, Yan pointed out. It is rolling out compatibility for domestic chips alongside its M3 and H3 models, with a large domestic computing cluster set to go online soon, while also aiming to further cut per-token inference costs, he said.

“The target for M3.1 is to bring inference costs down to roughly one-third of the level when M3 first launched,” Yan stressed. Cost reductions will enable price adjustments that will lower usage barriers, bringing in more users and larger token volumes, likely leading to continued improvement in the company’s gross margins, he added.

Source: Yicai Global

 

---