Alibaba Plans Revenue Sharing for Major Commercial Users of Its Open-Source Qwen AI Model

Alibaba's latest AI model puts it back in the great game

The ecommerce and cloud company hopes to recoup some of its lost shine with Monday's release Chinese tech companies are light years ahead of their US peers when it comes to making investors feel disenchanted. Nine years ago, Alibaba and Amazon both sported market capitalisations just shy of $500bn. Today, Amazon is worth $3tn; Alibaba a tenth of that. The Hangzhou-based ecommerce and cloud company hopes to recoup some of its lost shine with Monday's release of its latest AI model, Qwen 3.8-Max. Alibaba is making big claims about its capabilities. Early benchmarking suggests it's good -- if not the best -- in some areas such as "agentic coding", where bots write and fix code unassisted. To be clear, Alibaba's woeful share performance owed less to a perceived AI deficiency and more to political troubles. At home, Beijing spent years reining in the company and fellow tech titan Tencent. Overseas, the US restricts China's access to advanced chips and other technology. Last month, the EU handed Alibaba a hefty fine for failures on its online marketplace. And like its peers in the US, Alibaba is spending liberally. While nothing like the outsized bills of its US peers, capital expenditure of $18.3bn in its last accounting year left the company with negative free cash flow. Inflows are expected to return this year, but won't surpass 2024 levels until 2031, according to Visible Alpha estimates. Alibaba can at least make a good case for spending money on its own AI. Its ecommerce operation usually throws off cash, and its fast-growing cloud computing division gives it more opportunities to monetise its models. Like Microsoft, Alibaba offers "model as a service", giving smaller companies cloud-based access to pre-trained models, which they can then deploy, as chatbots perhaps, paying according to their usage. Counter-intuitively, the bifurcated tech world that splits China from the US can also work to China's advantage. It has forced innovation: like Google, Alibaba has its own proprietary chips. And US hardware makers need local partners. Apple won approval to launch AI-embedded devices in China using local tech, including Qwen. Alibaba chief Eddie Wu thinks AI models and services will make up half of external cloud income by 2027 -- about a tenth of total sales, Lex reckons. That's not huge but it is a useful launch pad. Qwen is open weight, so users can't totally replicate it, making it more commercial than previous open-source models. The race isn't just to make top-of-the-range AI, but to make AI that pays its way; don't count Alibaba out on that score.

EXCLUSIVE: Alibaba plans to charge big users of its next open-source AI model, sources say

Aug 7 (Reuters) - Chinese technology giant Alibaba (9988.HK), opens new tab plans to ask major users of the next version of its Qwen open-source AI model for a share of revenue they make from the hotly anticipated offering, according to two people familiar with the company's plans. Like last month's blockbuster model release from Chinese AI startup Moonshot, Alibaba's Qwen3.8-Max model is an open-source, open-weight model, meaning that the underlying learned settings that allow developers to run or adapt the system are available for download. By contrast, U.S. developers OpenAI, Anthropic and Alphabet's (GOOGL.O), opens new tab Google have closed-source models. Open source is often thought of as free, but tucked into the licensing terms, opens new tab of Moonshot's Kimi K3 was a provision that requires anyone offering the model for sale as a service and generating more than $20 million in annual sales to work out a commercial agreement with Moonshot. Alibaba plans to implement a similar measure for its open-source model next week, said the two people familiar with its ⁠strategy, which has not been reported previously. The sources requested anonymity because the plans are not public. To date, Alibaba has charged developers for use of its models when hosted on its own cloud computing platform but allowed most of its open-source offerings to be used in customers' own data centers without payment. The plans show Chinese AI firms, which have shocked markets by releasing open-source models nearly as capable as those from OpenAI and Anthropic, are converging on a business model as they push to take market share from their U.S. rivals. REVENUE-SHARING DEALS Deals under which U.S. firms share revenue generated by Chinese models with the Chinese AI labs that created them are taking shape, even as the White House has accused Moonshot of stealing technology from Anthropic. Chinese officials have rejected that claim as unfounded. The Kimi K3 license requires Moonshot's partners to share revenue, according to the two people, with Moonshot requiring up to a 30% revenue share, one of them said. Alibaba also plans to ask for a revenue share, both people said, but the rate remains unclear as the discussions are ongoing. Moonshot did not ⁠respond to a request for comment. Chinasoft International (0354.HK), opens new tab, a Chinese IT services provider, last month disclosed a revenue-sharing agreement with Moonshot in a regulatory filing, without disclosing the percentage. The Chinese AI firms are following a common playbook from Silicon Valley: Offer software at low or zero cost to start, charging for heavy commercial use and additional services. "You pay for collaboration with these open-weight model labs to make sure that you're optimizing your deployment. You pay for getting early access for the next revision of the model," said Paddy Srinivasan, CEO of cloud computing firm DigitalOcean Holdings (DOCN.N), opens new tab, one of several U.S. firms offering ⁠Kimi K3 and other Chinese models. He confirmed his company has a commercial agreement with Moonshot, but declined to discuss specifics. "This is a tried and tested open-source 'freemium' model," Srinivasan said. Kimi K3 costs about a third of Anthropic's Fable model, according to listed prices of input and output tokens for both models. Dan Fu, vice president of kernels at Together AI, said companies like his that offer AI software make their ⁠money by optimizing the service, such as better use of tokens, the building blocks of AI queries. "At the application layer, there's value out there for how you use it, how you actually get the models and the tokens to do something useful," Fu said. Open-source AI, meanwhile, is expanding, with U.S. firms joining the fray. Thinking Machines Lab, a San Francisco-based AI startup ⁠founded last year by OpenAI's former Chief Technology Officer Mira Murati, last month released its first open-source model, and is widely expected to follow with more powerful ones. "I don't see a fundamental barrier" to powerful open-source U.S. models, said Lin Qiao, CEO and co-founder of Silicon Valley-based Fireworks AI, who declined to discuss the company's commercial arrangements with Moonshot. "We are really waiting for that to happen." Reporting by Stephen Nellis in San Francisco and Eduardo Baptista in Beijing; Editing by Peter Henderson and Jamie Freed Our Standards: The Thomson Reuters Trust Principles., opens new tab * Suggested Topics: * Artificial Intelligence Eduardo Baptista Thomson Reuters Eduardo Baptista is a Senior Correspondent for Reuters based in Beijing, covering China's technology, space, and automotive industries. He has led enterprise and investigative reporting on China's military-linked companies, artificial intelligence and semiconductor supply chains, as well as macroeconomic and industrial policy. Baptista has reported from China for nearly a decade and holds a BA in History from the University of Cambridge.

Alibaba wants to charge the biggest users of its 'open' AI model

The next Qwen model will stay open-weight, but the largest commercial users may have to pay, a sign China's labs are hunting for a way to make open source pay. Alibaba is trying to make its free AI model pay. According to people familiar with the plan, the company intends to charge the biggest commercial users of its next open-source Qwen model, a shift that is subtle but significant. The model itself would stay open-weight, so its underlying settings could still be freely downloaded and run by anyone who wanted them, yet the largest users, the ones building serious commercial products on top of it, would now have to negotiate a commercial agreement first. That marks a break from how things worked before, since Alibaba has generally let companies run its open models on their own infrastructure without paying, using openness to win over developers and spread its technology. What makes monetisation tempting is Qwen's success. Alibaba has pushed the family hard, unveiling Qwen3.8-Max as its most capable model and closing the gap on the largest rivals, and the models have earned real standing along the way. The company has claimed Qwen ranks as the world's number-two open-weight model, which gives it the kind of reach that makes charging heavy users plausible. The template comes from a rival. Alibaba is mirroring Moonshot, whose Kimi K3 model asks partners for revenue sharing once they cross a threshold, effectively taxing the biggest beneficiaries of its free release. Those terms are specific, too: the arrangement can require up to 30% revenue sharing for partners with more than $20m in annual sales, a real cost for large deployers while staying free for everyone else. Alibaba's own terms are less settled, because the company has not yet fixed its revenue-share percentage, and the details are reportedly still being worked out as it prepares the next release. The logic will be familiar to anyone who has followed software. It is essentially a freemium model, free for the many and paid for the few who make serious money from it, adapted here to the economics of large language models. It also answers a hard question, since open models are expensive to train but generate no direct revenue when given away, and Chinese labs have been hunting for a way to fund the next round of development. That search is unfolding against a fierce backdrop. A brutal price war has raged among China's AI firms, with DeepSeek making a 75% discount permanent and squeezing everyone's ability to charge for access. The compute bills, meanwhile, are enormous. Training these models takes vast fleets of chips, and Moonshot reportedly used 20,000 Nvidia chips from Alibaba's cloud, so someone eventually has to pay for that hardware. The move also reframes what "open" means. If the biggest users must sign a contract, the label describes the licence more than the price, which blurs the line between open source and a commercial product. There is a geopolitical edge to it as well, because China's labs have leaned on openness partly to spread their models worldwide while US leaders kept their best systems closed, and charging heavy users risks blunting that advantage. Even so, developers may not mind the trade. Smaller teams keep free access, while large enterprises are used to paying for support, tuning and reliability, so a paid tier for heavy commercial use may feel reasonable rather than a betrayal. The bigger test is whether openness can fund itself at all. If revenue sharing works, it offers a path for open models to keep pace with the enormous budgets behind closed ones, and it does so without simply giving the technology away for nothing in return. For Alibaba, then, this is a bet that reach can eventually be converted into revenue. Having spent heavily to make Qwen popular in the first place, it now wants to collect from the companies that have gone on to build real businesses on top of it.

Alibaba plans revenue sharing for next open-source Qwen AI model

Alibaba plans to require large commercial users of the open-weight version of its Qwen3.8-Max AI model to share a portion of the revenue they generate from it, according to Reuters, citing two people familiar with the company's plans. The company intends to roll out the measure alongside the open-weight release of Qwen3.8-Max, which Reuters reports is expected next week. The specific revenue-share rate has not been finalized, as negotiations are ongoing. The move follows a similar licensing approach taken by Chinese AI startup Moonshot for its Kimi K3 model. Moonshot's terms require any party selling Kimi K3 as a service and generating more than $20 million in annual revenue to reach a commercial agreement with Moonshot. That agreement can include a revenue share of up to 30%, Reuters reports, citing one of the people familiar with the matter. Chinasoft International disclosed a revenue-sharing agreement with Moonshot in a regulatory filing last month, without specifying a percentage. Until now, Alibaba has charged developers for using its models through its own cloud platform, while allowing most open-weight deployments in customers' own data centers without a fee. If implemented, the change would bring revenue expectations to commercial deployments that currently operate beyond the reach of Alibaba Cloud. Alibaba released Qwen3.8-Max earlier this week, describing it as its most capable model to date, with 2.4 trillion total parameters and 95 billion active parameters. The release marked the first time Alibaba has open-sourced a model at this scale, with open weights planned for next week. Alibaba stock rose 4.5% in premarket trading in New York and 7% on the Hong Kong exchange following the announcement. DigitalOcean CEO Paddy Srinivasan, whose company is among several U.S. firms that carry Kimi K3 and other Chinese AI models, acknowledged that DigitalOcean has struck a commercial arrangement with Moonshot, though he would not elaborate on its terms. "This is a tried and tested open-source 'freemium' model," Srinivasan said. Dan Fu, vice president of kernels at Together AI, said that AI software providers find their revenue in the efficiency gains they deliver around tokens -- the fundamental units that AI systems use to process queries. "At the application layer, there's value out there for how you use it, how you actually get the models and the tokens to do something useful," Fu said.

Exclusive-Alibaba Plans to Charge Big Users of Its Next Open-Source AI Model, Sources Say

By Stephen Nellis and Eduardo Baptista Aug 7 (Reuters) - Chinese technology giant Alibaba plans to ask major users ⁠of ⁠the next version of its Qwen open-source AI model for ⁠a share of revenue they make from the hotly anticipated offering, according to two people familiar with the company's plans. Like last month's blockbuster model release from Chinese AI startup Moonshot, Alibaba's Qwen3.8-Max model is an open-source, open-weight model, meaning that the underlying learned settings that allow developers to run or adapt the system are available for download. By contrast, U.S. developers OpenAI, Anthropic and Alphabet's Google have closed-source models. Open source is often thought of as free, but tucked into the licensing terms of Moonshot's Kimi K3 was a provision that ⁠requires anyone offering the ⁠model for sale as a service and generating more than $20 million in annual sales to work out a commercial agreement with Moonshot. Alibaba plans to implement a similar measure for its open-source model next week, said the two people familiar with its strategy, which has not been reported previously. The sources requested anonymity because the plans are not public. To date, Alibaba has charged developers for use of its models when hosted on its own cloud computing platform but allowed most of its open-source offerings to be used in customers' own data centers without payment. The plans show Chinese AI firms, which have shocked markets by releasing open-source models nearly as capable as ⁠those ⁠from OpenAI and Anthropic, are converging on ⁠a business model as they push to take market share from their U.S. rivals. REVENUE-SHARING DEALS Deals under which U.S. firms share revenue generated by Chinese models with the Chinese AI labs that created them are taking shape, even as the ⁠White House has accused Moonshot of stealing technology from Anthropic. Chinese officials have rejected that claim as unfounded. The Kimi K3 license requires Moonshot's partners to share revenue, according to the two people, with Moonshot requiring up to a 30% revenue share, one of them said. Alibaba also plans to ask for a revenue share, both people said, but the rate remains unclear as the discussions are ongoing. Moonshot did not respond to a request for comment. Chinasoft International, a Chinese IT services provider, last month disclosed a revenue-sharing agreement with Moonshot in a regulatory ⁠filing, without disclosing the percentage. The Chinese AI firms are following a common playbook from Silicon Valley: Offer software at low or zero cost ⁠to start, charging for heavy commercial use and additional services. "You pay for collaboration with these open-weight model labs to make sure that you're optimizing your deployment. You pay for getting early access for the next revision of the model," said Paddy Srinivasan, CEO of cloud computing firm DigitalOcean Holdings, one of several U.S. firms offering Kimi K3 and other Chinese models. He confirmed his company has a commercial agreement with Moonshot, but declined to discuss specifics. "This is a tried and tested open-source 'freemium' model," Srinivasan said. Kimi K3 costs about a third of Anthropic's Fable model, according to listed prices of input and output tokens for both models. Dan Fu, vice president of kernels at Together AI, said companies like his that offer AI software make their money by optimizing the service, such as better use of tokens, the building blocks of AI queries. "At the application layer, there's value out there for how you use it, how you actually get ⁠the models and the tokens to do something useful," Fu said. Open-source AI, meanwhile, is expanding, with U.S. firms joining the fray. Thinking Machines Lab, a San Francisco-based AI startup founded last year by OpenAI's former Chief Technology Officer Mira Murati, last month released its first open-source model, and is widely expected to follow with more powerful ones. "I don't see a fundamental barrier" to powerful open-source U.S. models, said Lin Qiao, CEO and co-founder of Silicon Valley-based Fireworks AI, who declined to discuss the company's commercial arrangements with Moonshot. "We are really waiting for that to happen." (Reporting by Stephen Nellis in San Francisco and Eduardo Baptista in Beijing; Editing by Peter Henderson and Jamie Freed)

Alibaba open-source AI model: Alibaba plans to charge big users of its next open-source AI model

Like last month's blockbuster model release from Chinese AI startup Moonshot, Alibaba's Qwen3.8-Max model is an open-source, open-weight model, meaning that the underlying learned settings that allow developers to run or adapt the system are available for download. By contrast, U.S. developers OpenAI, Anthropic and Alphabet's Google have closed-source models. Chinese technology giant Alibaba plans to ask major users of the next version of its Qwen open-source AI model for a share of revenue they make from the hotly anticipated offering, according to two people familiar with the company's plans. Like last month's blockbuster model release from Chinese AI startup Moonshot, Alibaba's Qwen3.8-Max model is an open-source, open-weight model, meaning that the underlying learned settings that allow developers to run or adapt the system are available for download. By contrast, U.S. developers OpenAI, Anthropic and Alphabet's Google have closed-source models. Open source is often thought of as free, but tucked into the licensing terms of Moonshot's Kimi K3 was a provision that requires anyone offering the ⁠model for ⁠sale as a service and generating more than $20 million in annual sales to work out a commercial agreement with Moonshot. Alibaba plans to implement a similar measure for its open-source model next week, said the two people familiar with its strategy, which has not been reported previously. The sources requested anonymity because the plans are not public. To date, Alibaba has charged developers for use of its models when hosted on its own cloud computing platform but allowed most of its open-source offerings to be used in customers' own data centers without payment. The plans show Chinese AI firms, which have shocked markets by releasing open-source models nearly as capable as those from OpenAI and Anthropic, are converging on a business model as they push to take market share from their U.S. ⁠rivals. REVENUE-SHARING DEALS Deals under which U.S. firms share revenue generated by Chinese models with the Chinese AI labs that created them are taking shape, even as the White House has accused Moonshot of stealing technology from Anthropic. Chinese officials have rejected that claim as unfounded. The Kimi K3 ⁠license requires Moonshot's partners to share revenue, according to the two people, with Moonshot requiring up to a 30% revenue share, one of them said. Alibaba also plans to ask for a revenue share, both people said, but the rate remains unclear as the discussions are ongoing. Moonshot did not respond to a request for comment. Chinasoft International, a Chinese IT services provider, last month disclosed a revenue-sharing agreement with Moonshot in a regulatory filing, without disclosing the percentage. The Chinese AI firms are following a common playbook from Silicon Valley: Offer software at low or zero cost to start, charging for heavy commercial use and additional services. "You pay for collaboration with these open-weight model labs to make sure that you're optimizing your deployment. You pay for getting early access for the next revision of the model," said Paddy Srinivasan, CEO of cloud computing firm DigitalOcean Holdings, one of several U.S. firms offering Kimi K3 and other Chinese models. He confirmed his company has a commercial agreement with Moonshot, but declined to discuss specifics. "This is a tried and tested open-source 'freemium' model," Srinivasan said. Kimi K3 costs about a third of Anthropic's Fable model, according to listed prices ⁠of input and output tokens for both models. Dan Fu, vice president of kernels at Together AI, said companies like his that offer AI software make their money by optimizing the service, such as better use of tokens, the building blocks of AI queries. "At the application layer, there's value out there for how you use it, how you actually get the models and the tokens to do something useful," Fu said. Open-source AI, meanwhile, is expanding, with U.S. firms joining the fray. Thinking Machines Lab, a San Francisco-based AI startup founded last year by OpenAI's former Chief Technology Officer Mira Murati, last month released its first open-source model, and is widely expected to follow with more powerful ones. "I don't see a fundamental barrier" to powerful open-source U.S. models, said Lin Qiao, CEO and co-founder of Silicon Valley-based Fireworks AI, who declined to discuss the company's commercial arrangements with Moonshot. "We are really waiting for that to happen."

Alibaba's Next Qwen AI Model May No Longer Be Free for Large Enterprises as It Eyes Moonshot Kimi K3-Like

Alibaba Reportedly Plans Revenue Sharing for Qwen 3.8-Max Alibaba plans to require major commercial users of the next version of its open-source AI model, Qwen3.8-Max, to share a portion of the revenue they generate from the model, Reuters reported on Thursday, citing two people familiar with the plans. The model, expected to launch next week, will reportedly follow a licensing approach similar to Chinese AI startup Moonshot's Kimi K3. While the model's weights will remain freely available for developers to download, enterprises generating significant revenue from commercial AI services may be required to negotiate a revenue-sharing agreement with Alibaba. The exact percentage Alibaba intends to seek has not been finalized, according to the report. Chinese AI Firms Shift Toward Freemium Business Model Moonshot reportedly introduced a similar framework with Kimi K3 last month. Under its licensing terms, companies that generate more than $20 million in annual sales from services built on the model must enter into a commercial agreement with Moonshot. The publication, citing one source, reported that Moonshot can seek revenue sharing of up to 30%. Until now, Alibaba has primarily generated revenue by charging customers who access its AI models through Alibaba Cloud while allowing most developers to run open-source versions on their own infrastructure without additional licensing fees. The reported policy change suggests Chinese AI developers are increasingly adopting a "freemium" strategy -- offering open-source models to encourage widespread adoption while monetizing large-scale commercial deployments. Alibaba and Moonshot AI did not immediately respond to Benzinga's request for comment. Alibaba Tests New AI Monetization Model While proprietary AI providers charge customers via API access, companies like Alibaba and Moonshot appear to preserve open access while seeking a share of revenue from enterprises that build profitable businesses on top of their models. Industry executives told the publication that the approach mirrors longstanding open-source software business models, where enterprise customers pay for commercial rights, support and early access despite the underlying software remaining openly available. Price Action: Shares of Alibaba closed at $126.81 on Thursday, down 1.34%, while after-hours trading saw the stock edge lower to $126.79, according to Benzinga Pro. According to Benzinga Edge Stock Rankings, Alibaba ranks in the 90th percentile for Value, although the stock has faced short- and medium-term weakness while maintaining positive long-term performance. Disclaimer: This content was partially produced with the help of AI tools and was reviewed and published by Benzinga editors. Photo courtesy: Shutterstock Market News and Data brought to you by Benzinga APIs To add Benzinga News as your preferred source on Google, click here.

Alibaba may start charging for use of latest AI model

STORY: :: Stephen Nellis, Technology Correspondent So why might Alibaba, the Chinese tech giant that's taken the world by storm, by offering a free download of its cutting edge AI, want to start charging money just like its rivals in the United States? :: San Francisco, California / August 6, 2026 Like all major AI companies, Alibaba spends a ton of money training and developing its AI models. So like all companies, Alibaba wants to recoup some of that cost. Now they've always charged money if you want to buy it directly from Ali Cloud, but if you wanted to run it in your own data center, you could do that without necessarily paying Alibaba. That might change as soon as next week. The U.S. labs are largely what's called closed source. That means you can only really access their models by going directly to them and paying them to get it from their data centers. Now, companies like Alibaba offer it on that basis too, but you can also download the model and run it in your own data center without necessarily having to pay them. That is changing. A few weeks ago, Moonshot, the creator of the Kimi K3 model, put out a model that you can run in your own data center. But if you make more than $20 million a year, you need to work at a commercial agreement with them that we understand from our sources involves sharing quite a bit of your revenue. We understand from those same sources that Alibaba is poised to do the same thing as soon as next week. One of the questions this raises is whether this new strategy from the Chinese AI companies is trouble for U.S. companies like OpenAI. In the short term, probably so. These Chinese companies have been able to offer extremely competitive technology in terms of what the models can do at a much lower price, sometimes at about a third of the price of US models. However, these models are still very expensive to operate. So over the long term, most of the sources in the AI community think that these prices will start to come closer together and it'll be a much more vigorous competition between the two countries and the labs in each.