What matters in AI.

Subscribe

Alibaba says Qwen4-Max has a score above that of Claude Fable 5.1

Alibaba gave no scores and kept the weights of the model closed. No external group can check the claim.

Alibaba launched Qwen4-Max, its new flagship AI model. Alibaba is a Chinese company for online stores that also makes AI. It says the model received scores near the top of coding benchmarks, which are standard tests. It says the model received a score above that of Claude Fable 5.1 from Anthropic and Kimi K3.1 Pro from Moonshot.

What is new

Qwen4-Max is the top model of the Qwen series. Alibaba did not make it open weight. Open weight is when persons can get the model files and change them. Alibaba says small versions of the series will come.

The proof

Alibaba gave no scores, no price and no test by an external group. The claim of the position comes from Alibaba. Alibaba also says the model is as good as the models of Anthropic and OpenAI.

What is not known

It is not known which tests Alibaba used, or how large the difference is. It is not known when the small versions come, or if their weights are open.

Insight: Alibaba keeps the top model closed while a rival opens its model

Meituan, a Chinese food company, released the open weights of its LongCat 2.5 model with an MIT licence. This licence lets you use the files in your work. Alibaba holds back Qwen4-Max, and Qwen goes into Siri in China in December. This reads as a difference in method: Alibaba keeps the top model closed, and small versions are the next release. Look for open weights in the small versions.

Sources

  1. Alibaba Unveils New Flagship AI Model in Riposte to US RivalsBloomberg
AI MATTER · NEWS · AI MATTER · NEWS ·16 NOV2026

Posted

Tags

Learn the terms in this story

Guides for this story

More in Models

All Models news