From Wuhan to Silicon Valley, Manus took nine months

Author: Wawa, TechFlow

Today's biggest news in the AI world: Meta acquires Manus for several billion dollars.

This is Meta's third-largest acquisition in history, behind only WhatsApp and Scale AI, and more expensive than its purchase of Instagram.

Looking at Manus' timeline, the product launched in March this year and was acquired by December. From launch to sale: just nine months.

The founder, Xiao Hong, is from Ji'an, Jiangxi Province, a graduate of Huazhong University of Science and Technology, who started his entrepreneurial journey in Wuhan. His first product was Yiban, a WeChat official account formatting tool, which he sold. The second was Weiban, a CRM tool for Enterprise WeChat, also sold. The third was Monica, a browser AI extension—not sold, but heavily criticized.

Criticized for what? Being a shell.

At that time, the industry consensus was that only companies building large models had prospects; anyone building applications on top of others' models was merely a shell with no technical depth.

When Manus first gained attention in March this year, co-founder Ji Yichao replied to a user question on social media: "We've used Claude, and fine-tuned versions of Qwen (Tongyi) models."

Meaning they used other people's large models and built their own application layer on top.

So what?

Now it's worth several billion dollars.

Last year, senior ByteDance executives flew specially to Hong Kong to meet Xiao Hong, offering $30 million to acquire the company. He declined.

In hindsight, the difference between $30 million and several billion isn't just one year—it’s:

One product actually being built.

Besides, we think the most interesting part of this story isn't the ending, but the process.

In July this year, Manus made a decision: move the entire company from China to Singapore. Of the 120-person team, only 40 core engineers moved; the rest were laid off. The Beijing office closed, as did the Wuhan office.

Back then, many accused them of "run" (emigrating).

Now it’s clear: this step was necessary. A Chinese company being acquired by a U.S. tech giant would almost certainly fail regulatory approval under current conditions. Changing the registration location removed the obstacle.

Negotiations took only 10 days.

Liu Yuan, partner at ZhenFund, said it was so fast they initially suspected it might be a fake offer.

Ten days to finalize a multi-billion-dollar deal—how desperate is Meta?

Looking at the background: this year, Meta spent over $70 billion on AI capital expenditures, mostly on infrastructure, but has few usable products to show. OpenAI has ChatGPT, Google has Gemini—what does Meta have?

Llama is open-source and available to all. Meta needs a strong application-layer product—and Manus is exactly that, ready-made.

$125 million annualized revenue, built from zero in eight months, global users, subscription-based, proven viable.

This isn’t just acquiring a team—it’s acquiring a:

Proven business model.

Interestingly, Manus’ investor list includes Sequoia China, Tencent, and ZhenFund. When these funds invested, the valuation was tens of millions of dollars. Now exiting, returns are dozens of times higher.

So here it is: Chinese VCs invest in a Chinese-founded company, the company relocates to Singapore, gets acquired by an American firm, and the Chinese VCs earn money from an American company.

This chain is even more Agent-like than Manus’ product itself.

After the acquisition, Xiao Hong becomes Vice President at Meta. A founder who started in Wuhan building WeChat formatting tools is now heading to Silicon Valley to report directly to Zuckerberg.

ZhenFund’s Liu Yuan put it well: "The era for China’s generation of young entrepreneurs has arrived."

But that statement may only be half true.

The era has indeed arrived—but the way it arrives is by moving the company away.