# Qwen has officially released Qwen4-Max, now also on OpenRouter, OpenCode Go, Vercel AI Gateway, Novita AI, You.com, Lebrel and Token Harbor



Alibaba



Qwen4-Max is the first model in the Qwen4 series, officially released on a new sparse MoE architecture with 2.4T backbone parameters, 71B active per token, native vision and a 1M-token context window. Qwen says it beats Qwen3.8-Max while using up to about half the tokens, with lower API prices, especially for cached input, and a lower average cost per task on its Work360 evaluation. It is also on Vercel AI Gateway and Novita AI. Vercel lists the model ID alibaba/qwen4-max at 25% below Qwen3.8-Max pricing, with up to 384K output tokens; that is a regular list-price comparison, not a limited-time discount. Novita says it has been available from day one. Neither post gives a unit price. You.com also added Qwen4-Max to its Inference Router on day one, with a 1M-token context window and native text, image and video input; no price was given. Qwen4-Max is also live on Lebrel, where the Qwen team invites developers to try it for agentic coding. It is also live on Token Harbor, which the Qwen team shared and invited people to try.


## Availability

Available on QwenCloud through the Token Plan or the API, and in QwenWork, Qoder, the Qwen app and QwenStudio. It is also live on OpenRouter and OpenCode Go, both at $1.50 per million input tokens and $4.50 per million output tokens, with cache hits at $0.075. It is also on Vercel AI Gateway and Novita AI. On Vercel the model ID is alibaba/qwen4-max, priced 25% below Qwen3.8-Max, with up to 384K output tokens; Novita says it has been available from day one. It is also on You.com's Inference Router from day one. It is also live on Lebrel. It is also on Token Harbor \(qwen4-max\) at $1.50 per million input tokens, $4.50 output and $0.075 cached, using pay-as-you-go balance or an Office or Frontier Pass.
- Updated: Nov 18, 2026 20:05 · Beijing Time (UTC+8)
- Launched: Nov 16, 2026 10:01 · Beijing Time (UTC+8)

## Sources

- Source excerpt: https://x.com/qoder_ai_ide/status/2122165564420530599
- Source archive: https://qwen.ai/blog?id=qwen4-max
- Source excerpt: https://x.com/OpenRouter/status/2122158038631281136
- Source excerpt: https://x.com/OpenRouter/status/2122158050526273743
- Source excerpt: https://x.com/opencode/status/2122159680462163993
- Source archive: https://opencode.ai/docs/go/
- Source excerpt: https://x.com/Alibaba_Qwen/status/2122248112479907848
- Source excerpt: https://x.com/Alibaba_Qwen/status/2122286761883959700

Canonical entry: https://www.ailyre.com/source/?id=alibaba-qwen4-max&lang=en
