American AI is expensive. Some startups are turning to cheap Chinese models

npr ^ | July 15, 2026 | John Ruwitch

[BenLurkin]

San Francisco-based startup, Lindy.ai, creates artificial intelligence "assistants" to manage your email and calendar. At first, the company leaned heavily on Anthropic's top-of-the-line AI models.

But in meeting after meeting with his finance guy, Crivello said, one thing became clear: "By far, our No. 1 expense was Anthropic," he said. "Like, more than payroll."

More than payroll — for over two dozen employees. More than rent. More than for anything else. So last month, Crivello announced that Lindy had migrated 100% of its traffic to the Chinese AI model DeepSeek-V4.

"It was just 10x cheaper," he said, adding that it had saved the company millions of dollars. "So it was a very, very simple business decision."

Artificial intelligence has become one of the — it not the — fastest-growing costs for U.S. businesses. But for many companies, it's a double-edged sword: necessary but expensive. To survive, a growing number of firms are switching from American models to cheaper Chinese AI.

In the race to create the best AI models, U.S. companies like Anthropic, OpenAI and Google lead the world. Experts say Chinese models are six to 12 months behind in terms of capabilities.

But China has carved out a niche in open-source models, which are free to download and adapt. "The open-source scene right now is absolutely dominated by the Chinese. It's not even close," Crivello said.

He said every founder he knows who is working in the AI space either is thinking about switching to Chinese models or has done so already.

And ballooning AI costs are not just a startup issue either. Uber CEO Dara Khosrowshahi spoke about it last month on the Invest Like the Best podcast. "We blew through our AI budget in a quarter, you know, for the whole year, essentially. And it is forcing us to adjust," he said.

(Uber did not respond to NPR's request for information about whether it uses Chinese models.)

Bloomberg reported Airbnb CEO Brian Chesky as saying that last year the company relied on Alibaba's Qwen model, which was "good," "fast and cheap." Perplexity and Nvidia have also made use of Qwen.

Like a Ferrari or a Honda

Many companies are wary of trumpeting their use of Chinese models due to political sensitivities, but the models are widely available on AI-model hubs like Hugging Face, on the code hosting platform GitHub and via model aggregators and inference providers based outside China.

That includes the San Francisco-based company Featherless, which offers access to some 30,000 AI models. Founder and CEO Eugene Cheah said Chinese models are popular, even if they aren't "frontier" models, or best in class.

"It's like the difference between driving a Ferrari and a Honda. You can have the best luxury car, or you can just have a Honda at scale that works," he said.

"Actually, a lot of open-source AI groups are perfectly fine being N-1, N being where the frontier is," he continued. "Because as the gap keeps shrinking, at some point the question is: Does it actually matter?"

For many, like Lindy, it doesn't matter. The Honda of AI is perfectly good.

OpenRouter, another platform where startups can access a range of AI models, reported that use of China's DeepSeek has gone from around 9% to nearly 20% since January. Use of models from the Chinese companies MiniMax, Xiaomi and Tencent have also risen.

Some users download and self-host open-source Chinese AI models, but many use them via paid AI-hosting companies, like Featherless and OpenRouter, so that user data is kept in the United States.

Victor Su-Ortiz, who does global product marketing at the Shanghai-based MiniMax, attended a recent AI engineers conference in San Francisco. Companies pay to use AI models by paying for tokens, or units of AI work. Su-Ortiz said it all comes down to the cost per token.

"A lot of repetitive tasks can be done with a model that's just as performant but has much lower cost per token" when compared with leading AI models, he said. "And this is essentially what has brought these open-weight models into the United States."

He said companies are shifting from "tokenmaxxing" — using as much AI as possible — to saving costs by limiting usage, switching to cheaper models or routing different types of AI work to different kinds of models.

For research or "deep reasoning," for instance, the cutting-edge models may perform better, said Su-Ortiz. "But if you're routing for a coding task that is repetitive, high volume … that's where one of our models, especially MiniMax M3, will perform exceptionally well at like only one-tenth the cost."

Saving a few dollars isn't worth it for everyone For some companies, Chinese models still aren't good enough. Jon Gordner is CEO and co-founder of Comment.io, which was founded just weeks ago and is developing a product that he said is like Google Docs for coders and AI agents.

"We need to make as good software as we can as fast as possible. And for us, saving a few dollars on a cheaper model isn't worth it if we have to spend two or three more weeks fixing its mistakes," he said.

Gordner said his company is getting value out of Anthropic and OpenAI models in part because both companies are subsidizing users to hook customers. He said monthly subscriptions offer tokens at a huge discount now — but that probably won't last forever.

"Then for us, it's going to make a lot more sense to start evaluating Chinese models and open-source models," he said.

Ara Kharazian is lead economist at Ramp, a company that helps businesses track, control and automate spending. It has insight into AI spending, and Kharazian said he thinks U.S. companies will keep adapting — in other words, they may keep prices in check or introduce high-quality open-source models in a bid to outcompete Chinese rivals.

"The rise of these Chinese models is indicative of the fact that businesses want something that is today not being offered by the American model companies," he said. "The only reason why I'm bearish about the Chinese models is because I assume that the American model companies will respond competitively."

Gordner, of Comment.io, is less certain. He thinks the major U.S. AI companies may have to start charging more for AI as pressure to demonstrate profitability rises, possibly as they get closer to going public. Both Anthropic and OpenAI have filed confidential paperwork with the U.S. government to get the ball rolling on eventual initial public offerings.

"At some point," Gordner said, "the music's going to stop."

---

TOPICS: Business/Economy; Computers/Internet
KEYWORDS:

[SaveFerris]

(The rise of these Chinese models)

Ooh, let’s implement that awesome
Chinese Social Media Credit Score System

WHAT could possibly go wrong? 😕😲🤯😨

[butlerweave]

AI coded by Chinese , all your stuff will belong to them

[Oystir]

I believe apple phones for China have Qwen built-in

[bak3r]

Just my observation from clients I work with, everyone is switching over to on-site services running everything from tiny QWEN models to INT4 GLM5.2. Also before people get hyper about China, the biggest benefit of Chinese models is what I just stated, ON-SITE services.

[oldtech]

There’s always that danger. Better just not to involve China in much of anything. Are they not our enemy?

[butlerweave]

Things from China are not really cheaper it’s all subsidized by the government . Chinese EVs if the company goes out of business you’ll have a brick and not an EV

[rdcbn1]

In the tech world, it it’s cheap or free, it’s axiomatic that you are the product.

[alloysteel]

Definitely misplaced thrift. It costs much more to do it all over, than to do it right the first time.

But who knows what is right?

Good judgment comes from exercising a lot of bad judgment, then discarding what what does not work. Takes a LOT of time an treasure.

[Captain Compassion]

In the not too distant future AI will rule our lives. Do want your AI masters to be an American AI or a Chinese AI?

[Red Badger]

You get what you pay for................

[AnotherUnixGeek]

The most powerful and capable LLMs are always going to be hosted remotely and provided as a service we are charged for. By the time we can host the equivalent of Opus 4.8 on a future Mac Studio running at our houses, it'll be 2029 and the most powerful models will be far beyond what Opus 4.8 can do today.

Companies that can get useful work done with DeepSeek today can host it locally - DeepSeek is under the MIT open source license. That's why companies turn to it.

[Presbyterian Reporter]

Some of us oldsters remember the DotCom bust 26 years ago.

It looks like the techies are slow learners.

[SaveFerris]

(In the not-too-distant future AI will rule our lives.)

Ssshhhh! 🤫🤫🤫

We're supposed to pretend that can't happen!!

Geofencing: "Think of it as an invisible fence around you... that will be related to your face recognition, digital identity, and access control"

Big Tech whistleblower Aman Jabbi exposes the digital prison being constructed all around us in this clip from The Agenda: Their… pic.twitter.com/wrtYxorf4R— Oracle Films (@OracleFilmsUK) July 12, 2025

[ridesthemiles]

JUST WONDERING HOW ALL THESE POLITICIANS WILL COPE WHEN IT BECOMES VERY APPARENT THAT THEIR CHILDREN HAVE NO JOBS & HAVE BEEN REPLACED BY AI.

[Presbyterian Reporter]

“””JUST WONDERING HOW ALL THESE POLITICIANS WILL COPE WHEN IT BECOMES VERY APPARENT THAT THEIR CHILDREN HAVE NO JOBS & HAVE BEEN REPLACED BY AI.”””

---

I recall 30 plus years ago when Bill Gates at Microsoft was apolitical. Gates’ non-political stance ended when Clinton’s DOJ went after him.

Same thing will happen with AI, the politicians will demand that their children get cushy jobs at AI and if AI refuses to hire the children, the DOJ will go after them.

[SaveFerris]

(In the not-too-distant future)

He did a good job
cleaning up the place 🧼🧼🧼

But his bosses didn't like him
so they shot him into Space 🚀🌌🌌🌌

[Resolute Conservative]

We are hitting the top gear on this “AI for everything” stupidity.

[fireman15]

You get what you pay for................

Actually, with AI the best cliche’ is “Up is down, and down is up; north is south, and south is north”. Nothing about the current situation really makes any sense.

This is from someone who has spent a lot of time farting around with a wide variety of commercial services and also running models locally that are available in open-source forms.

The interesting thing is that the Chinese services and models currently have far less censorship built in than the models and services from the USA. A good example is if you try to create a political cartoon of a Democrat politician, in general none of the domestic services will do it. But if you want a political cartoon of a Republican or conservative they will all happily oblige.

As an example... You can have any of the US based image generators create anything you want about Lindsey Graham, but try to generate an image of Bill or Hillary or even James Carville??? Yeah... No...

[Scrambler Bob]

“By far, our No. 1 expense was Anthropic,” he said. “Like, more than payroll.”
= = =

So it costs more than the employees.

So maybe you can double, or more, the employees and forget AI.

Well, sorry, that assumes that any of the employees are competent.