On 2026-11-16 the official Qwen account announced the full launch of Qwen4-Max, following the Token Plan-only preview of November 11. Qwen says it is built on a new MoE architecture with native vision and a 1M-token context window, and that in internal evaluation it beats Qwen3.8-Max while using up to about 50% fewer tokens, with average cost per task down more than 60%. The API is live on QwenCloud at $1.5 per million input tokens, $4.5 per million output tokens and $0.075 per million for implicit caching, all below Qwen3.8-Max. It is also available in Qwen Studio, QwenWork and Qoder, and Qwen says open weights for smaller Qwen4 models will follow.

Key Takeaways

  • ✓Qwen4-Max moves from the November 11 Token Plan preview to a full launch, with the API now open on QwenCloud
  • ✓Per Qwen: new MoE architecture, native vision and 1M context; the QwenCloud model page lists 991K max input and 393K max output, with image and video input
  • ✓Efficiency: beats Qwen3.8-Max in internal evaluation with up to about 50% fewer tokens and more than 60% lower average cost per task (internal figures, no third-party benchmarks yet)
  • ✓Pricing: $1.5 input / $4.5 output / $0.075 implicit cache per million tokens, all below Qwen3.8-Max, with implicit caching 70% cheaper
  • ✓Compatible with the OpenAI and Anthropic APIs, with function calling, structured outputs, web search, batches and fine-tuning; Qwen says open weights for smaller Qwen4 models are coming
Alibaba Officially Launches Qwen4-Max: New MoE Architecture, Native Vision, 1M Context, API Live on QwenCloud at $1.5/$4.5 per Million Tokens
🖼️Official Media
Click to view high-res
🧭

Turn your technical choice into a development budget

Compare 40 dev plans & simulate token costs vs $20/mo subscriptions

🔬

In-Depth Technical Analysis

Background

On November 11 Alibaba released a Qwen4-Max preview, but only for Token Plan subscribers, so developers could not call it pay-as-you-go. On November 16 the official Qwen account announced the full launch: the API is live on QwenCloud, and the model is also in Qwen Studio, QwenWork and the AI coding tool Qoder. Qwen positions it as its flagship for work, creativity and discovery, focused on coding, office work, frontend design and complex agent tasks.

How it works

Qwen says Qwen4-Max uses a new MoE architecture with native vision and a 1M-token context window. The QwenCloud model page lists 991K max input, 393K max output and 262K max reasoning tokens, with text, image and video input. It supports prefix completion, function calling, context caching, structured outputs, batches, web search and fine-tuning, and is compatible with the OpenAI and Anthropic API formats.

Numbers that matter

The figures in the post come from internal evaluation: better results than Qwen3.8-Max with up to about 50% fewer tokens and more than 60% lower average cost per task. The post lists no specific benchmark scores, and there are no independent third-party results yet. Qwen says it handles real tasks in 3D modeling, game development, manufacturing, finance, law and healthcare.

Getting started

The API model name is qwen4-max. Pricing is $1.5 per million input tokens, $4.5 per million output tokens and $0.075 per million for implicit caching, with 5-minute explicit cache creation at $1.875, all below Qwen3.8-Max. Projects already using the OpenAI or Anthropic SDK can switch the base URL and model name to try it. Use caching to cut costs on long-context work. Qwen also says smaller Qwen4 models will get open weights.

Action HubReady to adopt this in production?

Benchmark side-by-side against alternatives, or calculate monthly token cost vs subscription break-even.

Up NextGoogle Launches Gemini 4 Hydrogen: Scores 51 on the Artificial Analysis Intelligence Index, Above Argon, Runs 2.5x Faster Than 3.8 Flash and Lifts Terminal-Bench 4.1 From 18.2% to 57%; $1.50/$7.50 per Million Tokens, Free Tier Included
Read Now→