Nvidia agrees to acquire Hugging Face for $12.93 billion

Nvidia has agreed to acquire Hugging Face, the widely used open-source AI model and dataset hosting platform, for approximately $12.93 billion, according to Nvidia CEO Jensen Huang's announcement. This is roughly 2.9 times Hugging Face's last known valuation of $4.5 billion from 2023, against reported annualised revenue of only about $150 million. No primary source (SEC filing, official press release, or deal documents) has been linked; reporting so far relies on secondary outlets covering the announcement.

If it closes as described, this hands Nvidia direct influence over the platform that much of the AI developer community relies on to publish, discover and download open models and datasets, extending its reach beyond chips into AI distribution infrastructure. That matters for developers and enterprises who depend on Hugging Face's neutrality, and for competitors like AMD, Google and others racing to reduce dependence on Nvidia hardware. The claimed "no Nvidia compute lock-in" commitment is the detail to watch, since it determines whether this is a genuine platform play or a strategic chokepoint dressed up as neutrality.

The framing that this is purely defensive, protecting Nvidia's hardware dominance against OpenAI, Anthropic and Google building their own chips, is speculation dressed as strategic analysis; no one has demonstrated this was Nvidia's actual motive. Coverage treating the deal as fully confirmed is premature given the absence of a primary filing or press release, and the story had already been trailed by Business Insider and The Information for over a week, so the "breaking news" framing overstates its novelty.

1. 01Nvidia's official 8-K or press release with deal terms, financing structure and expected closing date, since no primary source has surfaced yet
2. 02Whether antitrust regulators (US FTC/DOJ, EU Commission) open a review, given Nvidia's existing dominance in AI compute and this reaching into model/dataset distribution
3. 03Hugging Face's actual behaviour post-close: does it stay hardware-neutral (continue supporting AMD, Google TPU, etc.) or start favouring Nvidia-optimised deployments