✳ the wire · analysis

Qwen4-Max-Preview goes live in Qwen Chat and Alibaba's Model Studio Token Plan, as a two-week preview with a 1M-token context

Qwen 4 Maxconfirmedby ArtificialWatch

An email when Qwen 4 Max answers on a public API, about 15 minutes after members, plus every frontier launch · no card

ArtificialWatch title card · verified against a primary source

Qwen4-Max-Preview is out as a two-week preview. Alibaba's Qwen Chat lists it in its public model list (chat.qwen.ai/api/models), and Alibaba Cloud Model Studio offers it to developers through its Token Plan, a monthly credits subscription for individual developers. ArtificialWatch's radar caught it on Qwen Chat's list and sent an early-version alert at 02:11 UTC on November 11. ArtificialWatch read the Qwen Chat listing at 02:47 UTC and Model Studio's pages at 03:52 UTC.

WHAT ALIBABA'S PAGES SAY

  • Qwen Chat lists it as qwen4.0-max-preview. Model Studio also accepts qwen4-max-preview and bills both as qwen4.0-max-preview.
  • Context window: 1,000,000 tokens, per Qwen Chat's listing.
  • Capabilities: Qwen Chat flags vision, documents, video, audio, thinking and search; Model Studio lists reasoning, visual understanding and text generation.
  • The preview runs from 10:00 on November 11 to 24:00 on November 25, Beijing time (02:00 UTC November 11 to 16:00 UTC November 25). Model Studio says the model will be retired when the preview ends.
  • During the preview, Token Plan calls use as little as 20% of the standard credit rate, and 4% between 22:00 and 08:00 Beijing time, per Alibaba.
  • Alibaba's description pitches expert knowledge, complex reasoning, math, coding, long multi-step agent tasks with tool use, and front-end design. Those are Alibaba's own claims.
  • On November 10, Qwen Chat's front-end code had already made qwen4.0-max-preview the default model for agent mode.

WHAT IS NOT OUT

  • A pay-as-you-go price. Model Studio's pricing page has no Qwen 4 entry, so the Token Plan is the only paid route so far, and OpenRouter has no Qwen 4 id (read 03:53 UTC).
  • An announcement. Alibaba's Qwen account had not posted about it by 03:53 UTC.
  • The final Qwen 4 Max. This is a time-limited preview, so ArtificialWatch's main Qwen 4 Max alert is still armed for the final model.

WHAT WE DO NOT KNOW

When the final Qwen 4 Max ships and at what price, how the preview differs from it, and when the smaller Qwen 4 models (Plus, Flash and an open-weight 27B, all unconfirmed) follow. Alibaba has given no date for the final model; the preview ends November 25. Alibaba Cloud's CTO said 'Qwen4 next week' on October 29, per one attendee; the preview arrived 13 days later.

Source: Alibaba Cloud Model Studio ↗ · Qwen 4 Max tracker · the bench index
Data: AI model retirement calendar · Open-weights leaderboard · LLM context windows

Cite this page

Writing about this story? Copy a link or a citation.

<a href="https://artificialwatch.com/wire/qwen4-max-preview-live-in-qwen-chat-and-model-studio-token-plan">Qwen4-Max-Preview goes live in Qwen Chat and Alibaba's Model Studio Token Plan, as a two-week preview with a 1M-token context</a>
ArtificialWatch, "Qwen4-Max-Preview goes live in Qwen Chat and Alibaba's Model Studio Token Plan, as a two-week preview with a 1M-token context", https://artificialwatch.com/wire/qwen4-max-preview-live-in-qwen-chat-and-model-studio-token-plan
sweeping every 60 seconds

Hear when Qwen 4 Max drops — not when we write it up.

Claude Opus 5 went live at 16:51 UTC. Members had the alert at 16:52. Free gets a Chrome push and an email about 15 minutes after members.

  • Free forever
  • No card
  • Unsubscribe in one click

Want it the minute it answers? Try Signal 7 days free →

← back to the wire · every story, by month