Qwen Max 3.9 and Qwen Flash 3.9 Release Dates: What Alibaba Has Said

As of 16 October 2026, Alibaba has not released a Qwen Max 3.9 or a Qwen Flash 3.9, and its Model Studio documentation, updated on 15 October 2026, still recommends qwen3.8-max and qwen3.8-flash and lists nothing newer in either line. At 07:33 UK time on 16 October 2026 the versus page showed 50p (50%) for a Qwen Max 3.9 or later by 31 October and 30.5p (30.5%) for a Qwen Flash 3.9 or later by the same date. A price is the market's estimate of a chance between 0p and 100p: a winning contract settles at 100p and a losing one at 0p, so you can lose the whole amount you put into a position.

Two live predictions cover this. When will Alibaba's Qwen Max reach version 3.9? asks by which date a qualifying Max model will be public, with the answers 31 August, 30 September and 31 October. Will Alibaba release Qwen Flash version 3.9 or later by 31 Oct? is a yes or no question about the Flash line.

What Alibaba has said about its next models

Qwen is the family of AI models made by Alibaba. Its press release of 3 August 2026 introduced Qwen3.8-Max as the most powerful model in the series to date, with 2.4 trillion parameters and a context window of up to 1 million tokens, and said developers could reach it through the Model Studio platform. A press release of 20 May 2026 had introduced Qwen3.7-Max, so the two Max launches were 75 days apart.

On 22 September 2026, at its Apsara Conference, Alibaba said in a press release that its next-generation model, Qwen 4, is in training, and set out a roadmap for Qwen 4.5 and Qwen 5 at 5 to 10 trillion parameters. It gave no release date for any of them, and it does not mention a 3.9 model or a Flash model.

The Model Studio page on text generation, last updated on 15 October 2026, recommends qwen3.8-max for the strongest reasoning and qwen3.8-flash as the lower-cost choice, and lists no 3.9 model. Its list of newly released models, last updated on 11 October 2026, shows no new Qwen model in its text generation category since qwen3.8-max-0902, an upgraded snapshot of qwen3.8-max dated 2 September 2026 that keeps the 3.8 name. Its later Qwen entries are Omni, audio and image models, and the entries of 3 and 8 October only add a cloud region for existing models.

The Flash line so far

Qwen3.8-Flash is the newest Flash model. The list of newly released models dates it to 26 August 2026 and calls it the latest multimodal model of the Qwen family, with a context window of one million tokens. The same list dates the earlier Flash models: qwen3.5-flash to 23 February 2026, qwen3.6-flash to 17 April 2026 and qwen3.7-flash to 21 July 2026. The gaps between those four Model Studio dates were 53, 95 and 36 days.

Two other models carry Flash in their name but cannot settle a 3.9 answer. Qwen3.8-Flash-Next, on its Hugging Face page (the repository was created on 24 August 2026), is called an experimental preview of the architecture that will underpin Qwen4, and it is numbered 3.8. Qwen3.8-Omni-Flash, launched on 18 September 2026 according to Qwen's research page, is also numbered 3.8 and is the kind of narrow variant the Flash rules set aside.

How the two predictions settle

The Qwen Max prediction counts a model only if its name contains both Qwen and Max, it is the flagship of Alibaba's core Max tier and its version is 3.9 or higher. A jump to a new generation, for example Qwen4-Max, would count. Any Qwen3.8-Max release does not, even a later stable, open-weight or dated snapshot, and nor does a model without Max in its name or a narrow variant for vision or coding. Anyone must be able to use the model: an open beta or a waitlist anyone can join counts, private access does not. Alibaba's own information decides, with credible reporting as a cross-check. The answers are dates, and a qualifying release by a date makes that answer and every later one right. The 31 August and 30 September dates have passed with no qualifying release, which leaves 31 October. The market closes at 03:59 UTC on 1 November 2026, which is 03:59 in UK time (GMT, as the clocks go back on 25 October).

The Qwen Flash prediction works along the same lines for one yes or no answer. It looks for Alibaba's next core Flash flagship, with Qwen and Flash both in the name and a number of 3.9 or higher, reaching the public from 11:25 pm UTC on 31 August 2026 (00:25 on 1 September in UK time, BST) up to its cut-off. A Qwen3.9-Flash or a Qwen4-Flash would count. A Flash model still numbered 3.8 would not, and nor would a Qwen3.9-Max or any other model without Flash in its name. As with Max, an open beta or an open waitlist counts and private access does not, and Alibaba's information decides, with consistent press reports as a check. The rules give the cut-off as 11:59 pm Eastern Time on 31 October, which is 03:59 on 1 November in UK time (GMT). The market lists its end as 23:59 UTC on 31 October (23:59 UK time, GMT), four hours earlier, so read the prediction's page before you trade.

Dates and signals to watch

• Any day: a new Qwen entry on Model Studio's list of newly released models, a post on Qwen's research page or a new Qwen repository on Hugging Face. A public Qwen3.9-Max, Qwen3.9-Flash, Qwen4-Max or Qwen4-Flash would count for its own prediction. Another 3.8 snapshot, or an Omni, Plus or image model, would not.
• Day counts, not a schedule: 31 October is 66 days after the Model Studio date of Qwen3.8-Flash (26 August) and 89 days after the announcement of Qwen3.8-Max (3 August). Of the three earlier Flash gaps, 53 and 36 days were shorter than 66 and 95 was longer; the one Max gap in this guide, 75 days, was shorter than 89. Alibaba has named no date for either model, and the next gap could be shorter or longer.
• 31 October and 1 November 2026: the last open Max answer is dated 31 October, and the Max market closes at 03:59 UK time on 1 November, the same moment as the Flash cut-off under its Eastern Time rule. With no qualifying release by then, a contract on the 31 October Max answer and a contract on Flash Yes would each settle at 0p.

Before you trade

The rules, the price and the settling source are on each prediction's page before you trade, and how it works follows one position from opening to settlement. versus adds a 2% transaction fee on top of the stake and shows it before you confirm, and while a market is open you can sell at the current price. Safer Play tools, including deposit limits, loss limits, reality checks, time out and self exclusion, sit alongside GAMSTOP.

versus is licensed by the UK Gambling Commission under account 101143, which you can check on the Commission's public register. Positions settle in ordinary money, not cryptocurrency. This guide is for readers in Great Britain. Our guides to the Kimi K4 release date and the GLM 5.4 release date cover similar predictions.

Frequently asked questions

Is there a Qwen Max 3.9 or a Qwen Flash 3.9 yet?

No. As of 16 October 2026, Alibaba's Model Studio documentation lists qwen3.8-max and qwen3.8-flash as the newest, and its press room, Qwen's research page and Hugging Face organisation show no 3.9 model.

What is the latest Qwen Flash model?

Qwen3.8-Flash, which Model Studio's list of newly released models dates to 26 August 2026. Qwen3.8-Flash-Next is an experimental preview and Qwen3.8-Omni-Flash a specialist model, and both are numbered 3.8.

What is the latest Qwen Max model?

Qwen3.8-Max, which Alibaba announced on 3 August 2026. Model Studio also lists qwen3.8-max-0902, an upgraded snapshot dated 2 September 2026 that keeps the 3.8 name, so it does not count.

Would Qwen 4 count for these predictions?

A Qwen4-Max would count for the Max prediction and a Qwen4-Flash for the Flash prediction, if it is open to the public in time. Alibaba said on 22 September 2026 that Qwen 4 is in training and gave no date.

18+ only. Trade responsibly. BeGambleAware.org and GAMSTOP.