{"id":2243,"date":"2026-08-06T13:09:22","date_gmt":"2026-08-06T13:09:22","guid":{"rendered":"https:\/\/techvisor.pro\/?p=2243"},"modified":"2026-08-06T13:09:22","modified_gmt":"2026-08-06T13:09:22","slug":"qwen3-8-max-alibabas-new-model-with-2-4-trillion-parameters","status":"publish","type":"post","link":"https:\/\/techvisor.pro\/en\/qwen3-8-max-alibabas-new-model-with-2-4-trillion-parameters\/","title":{"rendered":"Qwen3.8-Max: Alibaba’s new model with 2.4 trillion parameters"},"content":{"rendered":"

On August 3, 2026, Alibaba released Qwen3.8-Max \u2014 the most powerful model in its Qwen lineup to date. We break down its specifications, benchmarks and why this is another step in the fast-moving race of large Chinese AI models.<\/p>\n

What Is Qwen3.8-Max<\/h2>\n

Qwen3.8-Max is Alibaba Cloud\u2019s flagship model with 2.4 trillion parameters, slightly smaller than Kimi K3 from Moonshot AI (2.8 trillion), which still remains the largest open model in the world. The model\u2019s context window is up to 1 million tokens, allowing it to process huge amounts of text, images and video in a single request.<\/p>\n

Architecture: Efficiency Through MoE<\/h2>\n

\"Architecture:<\/p>\n

The model is built on a sparse Mixture-of-Experts architecture \u2014 instead of activating all 2.4 trillion parameters at once, only about 95 billion are used for each request. This significantly reduces compute cost and response latency, allowing the model to process text, images and video without a proportional increase in expenses.<\/p>\n

Capabilities and Benchmarks<\/h2>\n

\"Capabilities<\/p>\n

According to Alibaba, Qwen3.8-Max delivers results comparable to Anthropic\u2019s Claude Fable 5, and in some cases better. The model ranked fifth in Text Arena and second in Vision Arena, trailing only Fable 5 in that category.<\/p>\n

Among the stated capabilities are application development, research work, legal document review, sports analytics, financial research and even architectural 3D modeling. Alibaba also claims that the model is capable of autonomously writing code for weeks with minimal human intervention.<\/p>\n

Open Weights \u2014 Next Week<\/h2>\n

\"Open<\/p>\n

Alibaba plans to publish the model\u2019s full weights within the week following release \u2014 this will be the company\u2019s first \u201cMax\u201d-class model to become fully open. For now, Qwen3.8-Max is available through the API on Alibaba Cloud Model Studio for developers worldwide.<\/p>\n

Context: The Race of Chinese Open Models Is Accelerating<\/h2>\n

The release of Qwen3.8-Max happened less than a month after Moonshot AI released Kimi K3 \u2014 a model the company called the world\u2019s first open \u201c3-trillion-class\u201d model. That same week, China\u2019s DeepSeek released the V4-Flash model, which researchers estimate could turn out to be one of the cheapest models in the world to run.<\/p>\n

Together, these releases show how quickly Chinese technology companies are increasing the pace of launching powerful open models, competing not only on size but also on deployment cost.<\/p>\n

In Brief: The Key Points<\/h2>\n