Alibaba Qwen Releases Qwen3.8-Max: A 2.4 Trillion Parameter MoE Model and the Most Capable One in the Qwen Family to Date
Alibaba released Qwen3.8-Max, a 2.4 trillion parameter Mixture-of-Experts model supporting multimodal inputs and a 1 million token context window.
Alibaba Cloud's Qwen division has transitioned its latest flagship model, Qwen3.8-Max, from preview to general availability. The system employs a Mixture-of-Experts architecture totaling 2.4 trillion parameters.
The model processes text, images, and video inputs simultaneously. It supports a context window of one million tokens, allowing for extensive data ingestion during inference tasks.
Developers can now access the model via published per-token pricing structures. The organization stated that open weights will become available to the public next week.
Although a formal benchmark table has not been released, this iteration marks the most capable entry in the Qwen family lineup so far.
This page provides an editorial summary based on publicly available information. It is not a republished article. Use the source link below for the original report.