Alibaba Cloud has officially deployed its latest flagship large language model, Qwen3.7-Max, to its integrated AI development platform, Bailian. This release marks a significant milestone in the company’s ‘Model-as-a-Service’ (MaaS) strategy, offering developers and enterprise clients immediate access via API and a streamlined subscription service known as the ‘Token Plan.’ The move is designed to lower the barrier for high-performance AI integration across China’s rapidly evolving digital economy.
The pricing structure for Qwen3.7-Max reflects a continued push toward the commoditization of advanced AI capabilities. Alibaba has set the input price at 12 RMB per million tokens and the output price at 36 RMB per million tokens. This aggressive pricing strategy is aimed at capturing a larger share of the enterprise market, particularly as competitors like Tencent and ByteDance’s Volcengine ramp up their own infrastructure and model offerings to win over corporate developers.
Beyond simple API access, the introduction of the Token Plan subscription model suggests a shift in how cloud providers monetize intelligence. By providing a predictable cost structure for heavy users, Alibaba is attempting to lock in ecosystem loyalty at a time when the daily revenue from token usage is reportedly growing at an exponential rate. This transition from experimental usage to industrial-scale deployment is becoming the new frontline in China’s tech sector.
This launch occurs against a backdrop of intensifying competition in the domestic AI landscape. Recent industry data indicates that Alibaba Cloud’s full-stack AI narrative is gaining significant traction, with some internal metrics showing daily token-related revenue increasing 15-fold over a five-month period. As the industry moves from a ‘war of a hundred models’ to a more consolidated market, the ability to deliver high-performance reasoning at scale and low cost will determine which cloud giant becomes the default operating system for the AI era.