Open-Source AI Models Are Quietly Taking Over - And the Frontier May Not Matter as Much as You Think

For much of this summer, the AI world had its eyes locked on Anthropic's latest frontier models and the political tug-of-war over who gets access to them. But while the spotlight stayed on the cutting edge, developers kept building - and they weren't waiting for permission from the big labs. The real action, it turns out, may be happening far from the frontier.

Chinese open-weight models accounted for 41% of downloads on Hugging Face this spring, beating out U.S. models. On OpenRouter, the six most popular models right now are all open models from Chinese companies like Tencent, Xiaomi, DeepSeek, MiniMax, and Z.ai - with Anthropic's Claude Opus 4.7 trailing in seventh. Data from Vercel shows open models now handle nearly a third of AI requests on its platform, while closed models serve as a pricier, premium layer. Those figures only capture part of the ecosystem, but they point to a growing shift: many companies are choosing cheaper, customizable alternatives over expensive frontier models.

Hugging Face CEO Clem Delangue says the trend reflects a deeper realization. "If you're an AI company, you don't want to outsource your core capabilities to a black box API you don't control," he said. Half of all Fortune 500 firms now use Hugging Face to deploy their own private or open-source models. Microsoft CEO Satya Nadella has also warned against single-provider lock-in, arguing that control of data should be a top priority. Meanwhile, the debate over open models' safety continues. Delangue argues that keeping models closed doesn't eliminate risk - it just concentrates power. "The biggest risk in AI is concentration of power," he said. "You make the world safer by leveling the playing field and creating transparency."

Source: TechCrunch