Within hours, the AI community had pieced together what was likely happening. The mystery model was widely believed to be Qwen3.8-Max undergoing pre-release testing, while its references to Anthropic's Claude looked less like intentional branding and more like the kind of training artifact that occasionally surfaces in large language models.

Alibaba says its latest model can compete with Anthropic’s Claude Fable 5.

Alibaba Group has unveiled its latest 2.4 trillion-parameter AI model Qwen3.8, placing it among the world's top-tier large language models in overall performance. Alibaba said Qwen3.8 trails only Anthropic's Claude models on Arena, a widely followed third-party AI evaluation platform.

The new open-weight model targets enterprise software engineering and multimodal workloads as enterprises weigh deployment efficiency alongside AI model performance.

Qwen3.8-Max is the Chinese e-commerce giant’s most capable LLM to date. It features 2.4 trillion parameters, about seven times more than the Qwen3.5 model that Alibaba released in February. The LLM activates 95 billion of its parameters when answering a query.

Alibaba Group Holding Ltd's latest move to scrap a planned initial public offering for its unit Cainiao Smart Logistics Network will help strengthen synergy between its logistics services and e-commerce businesses and expand its presence in overseas markets by increasing investment in global logistics networks, industry experts said.

Alibaba (BABA) said Monday in a social media post that it launched Qwen3.8-Max, its latest large language model, and plans to release the model's open weights next week alongside Qwen3.8-27B. The company said Qwen3.

Alibaba stock has gained momentum on its AI releases. That's not great for American model developers such as OpenAI and anthropic.