Skip to main content
AI RADARModels
2•11/16/2026, 08:47•1 min read

Qwen4-Max released: 50% fewer tokens and lower per-task cost

Qwen4-Max released: 50% fewer tokens and lower per-task cost - 1
Qwen4-Max released: 50% fewer tokens and lower per-task cost - 2
Qwen4-Max released: 50% fewer tokens and lower per-task cost - 3
Executive TL;DR30s read

Qwen4-Max introduces a MoE architecture, a 1M-token context window, built-in image support, and significantly lower token usage and cost.

Qwen4-Max is now available. The model uses 50% fewer tokens and is more than 60% cheaper per completed task, according to the announcement.

Key features:

  • 50% lower token usage.
  • More than 60% cost reduction per task.
  • Mixture-of-Experts (MoE) architecture.
  • 1 million token context window.
  • Built-in image support.

Qwen4-Max is positioned not only for programming, but also for practical workloads such as 3D modeling, manufacturing, finance, law, medicine, and scientific research.

The API is already available through the official Qwen4-Max page. Open weights for Qwen4-27B are planned for a later release.

Why it matters

Analysis

It signals a push toward cheaper, long-context multimodal models for real-world development and enterprise tasks.

Discuss in community

Share your questions and insights with developers

Join Telegram
+5 Points
← Back to news feed

Related in Ecosystem

Practical guides, core terminology, and agent skills