Alibaba (Ali) has released numerous large models, not simply a matter of “volume chasing,” but a carefully planned “Model-as-a-Service” (MaaS) ecosystem strategy. There are multiple considerations behind this, which can be summarized as “internal empowerment and external ecosystem building.”
Internal Business Driven (Inward Empowerment)
Alibaba possesses an extremely vast and diverse business landscape, including e-commerce (Taobao & Tmall), finance (Ant Financial), logistics (Cainiao), cloud computing (Aliyun), and entertainment (Youku), among others.
- Scenario Customization: A single, general-purpose large model cannot efficiently meet the fine-grained needs of all vertical scenarios. For example, the requirements for an e-commerce customer service model, an advertising creative generation model, a financial risk control model, and a logistics route planning model are entirely different.
- Efficiency & Cost: Training specialized “small models” for specific tasks (such as coding or diagramming) is more cost-effective and responsive than continuously calling upon a large “all-in-one” model.
- Therefore, Alibaba needs a “Model Matrix” with “Tongyi” (Tongyi) foundational model at its core, to derive specialized models tailored to different businesses, enabling an AI transformation of internal operations.
Building a Cloud Service Ecosystem (External Ecosystem Building)
This is Alibaba’s core strategy. Large Models are the “operating system” of the AI era, while computing power (cloud services) are like “water and electricity.” The goal of Alibaba Cloud is to become a leading “compute + model” infrastructure provider in the AI era.
-
Building a “Model Supermarket”: Alibaba not only provides its own “Tongyi” series models (such as Tongyi Qianwen, Tongyi Wanxiang, and Tongyi Lingma), but also gathers a large number of third-party and open-source models through the “MoDaBu”(ModelScope) community. By releasing numerous models, it’s intended to showcase its powerful model research capabilities and rich model library, attracting enterprises to move to the cloud.
-
“Open Source + Closed Source” Parallel Strategy: Alibaba’s (such as open-sourcing the Qwen series of “Tongyi Qianwen”) open-source strategy is aimed at rapidly building a developer community, capturing market share, accelerating technological iteration, and competing with international open-source forces like Meta’s Llama. After developers “practice” on open-source models, they are likely to choose Alibaba Cloud’s paid computing power and closed-source commercial models for final deployment.
Summary
In short, Alibaba’s release of numerous large models is characterized by a strategy of: building a core self-developed model matrix centered around the “Tongyu” series to deeply empower its vast and diverse internal businesses. Simultaneously, through “open source” and “Model as a Service” platforms, it’s constructing an open and thriving AI ecosystem on Alibaba Cloud, with the ultimate goal of driving growth in its core cloud computing business.