AnswersPublished14 sources
China’s AI Gap With the US Is Narrowing on Capability and Price
In Bloomberg’s website building test, Kimi K3 cost $11.91 and Alibaba’s Qwen 3.7 Max cost $4.08, compared with $48.99 for Anthropic’s Fable 5.
Published byEdited with GPT-6 LunaImages generated with GPT Image 2
Research answer
Key takeaways
- In Bloomberg’s website building test, Kimi K3 cost $11.91 and Alibaba’s Qwen 3.7 Max cost $4.08, compared with $48.99 for Anthropic’s Fable 5.
AI Prompt
openai.comCreate a landscape editorial hero image for this Studio Global article: What does Bloomberg Businessweek’s October 15 report reveal about China’s narrowing AI gap with the United States, including Moonshot’s July. Article summary: The report’s central message is that China is rapidly eroding America’s AI lead by offering near-frontier capabilities at much lower prices—making global adoption, not just benchmark supremacy, a decisive part of the com. Topic tags: general, news, general web, user generated. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts w
China’s AI models are closing the gap with US competitors not only in capability but also in cost. Bloomberg’s comparison highlights Moonshot’s Kimi K3 as approaching Anthropic’s most advanced model, while a website-building test shows how much the bill can differ for a single task. 2
10
Kimi K3 makes the capability gap harder to measure
Released in July, Moonshot’s Kimi K3 nearly matched the capabilities of Anthropic’s most advanced model in Bloomberg’s comparison, at a fraction of the cost. That is evidence of a narrowing gap—not proof that Chinese models lead across every task or benchmark. 2
10
The distinction matters for businesses. If a lower-cost model performs well enough for a specific job, buyers may weigh the price difference more heavily than a small performance gap. But results from one comparison cannot establish which model is best for a company’s full range of needs.
The website-building test shows the price difference
For a website-building task, Bloomberg Businessweek’s reported costs were $11.91 for Kimi K3 and $4.08 for Alibaba’s Qwen 3.7 Max. Anthropic’s Fable 5 cost $48.99 in a reported comparison of the same kind of task. 10
14
These are task costs, not a universal measure of model pricing or total business expense. They show that, in this test, the models produced a substantial cost spread. They do not tell buyers how each system will perform on different workloads, or whether the cheaper option will be adequate for a particular use.
A narrowing gap does not mean the race is settled
Gary Marcus, a New York University professor, has described the US-China AI contest as an “unwinnable” race, comparing it to Coke and Pepsi: two competitors can retain substantial positions without either eliminating the other. Bloomberg’s summary also says US models are generally better, while Chinese models are cheaper. 1
That view puts the model comparison in perspective. Kimi K3’s results and the website-building costs make a strong case that price and improving capability are central to China’s competitive challenge. They do not show that China has won the AI race—or that a single performance ranking can settle it.
Studio Global AI
Continue your research
This page includes a source-backed answer you can continue inside Studio Global.
People also ask
What is the short answer to "China’s AI Gap With the US Is Narrowing on Capability and Price"?
In Bloomberg’s website building test, Kimi K3 cost $11.91 and Alibaba’s Qwen 3.7 Max cost $4.08, compared with $48.99 for Anthropic’s Fable 5.