Qwen’s ¥1,499 Plan Is Selling a Workflow, Not a Chatbot

¥1,499 is a strange number to put next to a chatbot.

It is too much for casual curiosity, too little to disappear into a company software budget, and just large enough to make someone ask what they are actually buying.

Alibaba’s Qwen app is reportedly testing paid office-assistant plans that range from roughly ¥19 to ¥128 per month, with the highest annual plan reaching ¥1,499. Basic functions remain free, while video generation is charged through a separate quota system.

The interesting part is not that Qwen has added a paywall.

It is what the paywall is being placed around.

The product is moving away from the chat box

For a long time, consumer AI products were sold as open doors. You typed a question, received an answer, and decided whether the tool was useful enough to return to tomorrow.

That model encouraged enormous usage but not necessarily payment. A person might ask an AI to rewrite an email, explain a spreadsheet, or summarize a document without ever forming a stable relationship with the product.

An office assistant is different. It promises to sit inside a recurring rhythm: preparing a presentation before a meeting, turning notes into a document, analyzing a table, or helping with the small unfinished pieces of a workday.

The paid plan is therefore not only selling access to a stronger model. It is selling continuity.

That is a more delicate promise. A chatbot can be impressive once. An assistant has to become quietly dependable.

A subscription changes the meaning of “enough”

When a basic version remains free, the question is no longer whether people can use the product. They can.

The question becomes whether the free version feels deliberately incomplete.

This is not necessarily a bad strategy. Free access lowers the barrier for first-time users, while paid tiers can serve people who use documents, presentations, data analysis, or other heavier features often enough to notice limits.

But the difference between the tiers has to feel meaningful in the middle of an ordinary week. Not during a product demonstration. Not in a comparison chart with glowing icons. On a Tuesday afternoon, when the presentation needs another revision and the spreadsheet has one more ugly column than expected.

That is where paid AI products will be judged.

A monthly price also becomes difficult to compare when the units are different. One service may charge for model access, another for document processing, another for image or video credits. The lower price is not automatically the better deal if the feature someone needs is hidden behind a separate quota.

The annual ¥1,499 plan is roughly ¥125 per month when averaged across twelve months. That sounds more approachable than the headline figure, but only if the subscriber uses the office features often enough for them to become part of a routine.

Otherwise, it is simply another annual subscription waiting to be forgotten.

The Chinese AI market is learning how to charge for habit

This may be the more important story behind Qwen’s pricing.

China’s consumer AI products grew inside a culture of free access, aggressive distribution, and constant switching. Users could move between Qwen, Doubao, Kimi, Wenxin, and other assistants depending on the task or the mood. The model itself was often the product.

Paid office assistants suggest a different direction. The product is no longer only the model. It is the collection of small moments around the model: the document that opens in the right format, the presentation that does not need to be rebuilt from scratch, the analysis that remembers the previous file, the quota that lasts long enough to finish a real task.

This is also why direct “which AI is best?” comparisons are becoming less useful. The answer depends on what a person is repeatedly trying to do, how often they do it, and whether the paid plan covers the complete workflow rather than one attractive feature.

I am less interested in whether Qwen’s annual plan will immediately convert large numbers of casual users. I am more interested in whether it can become boring in the best possible way.

The successful office assistant may not be the one people praise most loudly. It may be the one they stop thinking about while they work.

For years, AI companies competed to make the first interaction feel remarkable. The next phase may be about making the fiftieth interaction feel worth paying for.

The free entrance to Qwen is still open. The new question is what happens when the work leads you further inside.