Official: Nvidia buys Hugging Face for $12.9 billion and takes over the largest AI library in the world

Exciting day for NVIDIA and @huggingface.

Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty. They allow every developer, startup, university, industry and country to build with, customize and benefit from AI.

Thank you…

— Jensen Huang (@JensenHuang) September 3, 2026

The agreement, confirmed jointly by Jensen HuangCEO of Nvidia, and Clément Delangue, CEO of Hugging Face, represents a drastic jump in the valuation of the New York startup, which in 2023 was listed at $4.5 billion. As part of the transaction, Nvidia will allocate additional funds to ensure the continuity of the platform’s operational and technical team.

The economic transaction was consolidated for an amount total of 12.9 billion dollarsto which is added a special talent retention fund of $1 billion to encourage key Hugging Face employees to stay after absorption. The figure reflects the dizzying value that the company has taken in the industry, supported by an unprecedented operational expansion within the artificial intelligence development ecosystem.

At the moment, The platform has an active base of 18 million registered developers and serves more than 200,000 enterprise customers worldwide. The repository houses about three million open source AI modelswhich demonstrates an exponential growth if one considers that In 2021 the library housed only 13,590 models before the massive emergence of this technology.

The purchase comes amid intense regulatory and business debate in Silicon Valley. As firms like OpenAI and Anthropic push for closed models for commercial security reasons, Nvidia and Hugging Face have led the defense of open source (open source), arguing that transparency accelerates innovation and guarantees the technological sovereignty of companies and institutions.

“Open models allow startups, companies, universities and public institutions to develop advanced capabilities without having to train each model from scratch,” said Jensen Huang. For his part, Delangue explained in an interview with CNBC that the decision to seek scale with Nvidia gained urgency after detecting the need to reinforce cybersecurity and provide free access to advanced tools in the face of the advance of corporate rental APIs.

With quarterly revenues that reached $96.22 billion and unprecedented cash flow, Nvidia has been deploying an aggressive investment and financing policy in the entrepreneurial ecosystem. In recent months, the company has committed more than $50 billion to AI development labs and supported the creation of a multibillion-dollar fund to build next-generation data centers.

Despite the size of the purchase, Nvidia confirmed to the US Securities and Exchange Commission (SEC) that Hugging Face will continue to operate independently and neutrally. The platform will continue to be compatible with hardware and software from other manufacturers, maintaining its premise of being an open and inclusive space for the entire global developer community.