Nvidia's Rumored Hugging Face Acquisition: Everything We Know Now
Nvidia has reportedly agreed to buy Hugging Face for $12.9 billion, according to a source familiar with the deal. Neither company has publicly confirmed the acquisition, and reports indicate that a final agreement could still fall through.
If the deal does go through, however, it would bring one of the world's most important open-model platforms under Nvidia's control.
Why Hugging Face?
Founded in 2016, Hugging Face has become one of the most popular hubs for developers to share and download open-source AI models and datasets. Its tools support everything from model development and optimization to inference and deployment.
But the $12.9 billion price tag is striking. Hugging Face's last major funding round, in 2023, valued the company at $4.5 billion. The company also reportedly turned down a $500 million investment offer from Nvidia last year that would have valued it at $7 billion.
Given that Hugging Face was recently generating about $150 million per year in revenue, it's still a relatively small fish in the AI pond. Hugging Face CEO Clément Delangue stated last month that the company is "close to profitability."
So why is Hugging Face worth so much to Nvidia?
The acquisition of Hugging Face could give the chip giant a strategic advantage in its bid to maintain dominance in the AI chip industry. The world's largest closed-source AI labs, including OpenAI, Anthropic, Google, and Amazon, are increasingly developing their own AI chips to reduce their reliance on Nvidia. At the same time, open models are giving developers more alternatives to proprietary AI platforms.
The Push for Open Models
As open models become more capable and widely used, the more demand there is for the GPUs used to train and run them. That's a good thing for Nvidia. In fact, the company has already spent billions of dollars developing its own open-source AI models.
Nvidia was also one of 25 American tech companies, including Hugging Face, that recently published a joint letter urging U.S. policymakers not to restrict open-weight models -- i.e. those that anyone can download, inspect, modify, and run on their own infrastructure.
The debate around the subject intensified earlier this summer when Chinese startup Moonshot AI released Kimi K3. It's an open-weight model that can outperform leading American proprietary models on some industry benchmarks while costing significantly less to run.
In the letter, the group of tech companies wrote that open models strengthen competition, which "keeps the gains of AI broadly shared rather than concentrated in a few hands."
The Democratization of AI
That's the promise of open models: democratizing access to increasingly powerful AI. Developers can simply download an open model, customize it for their needs, and run it themselves.
If Nvidia's reported acquisition of Hugging Face goes through, it's arguably a sign that a more open AI ecosystem is an integral part of the industry's future.
But democratizing access to models creates a new challenge: access to the compute needed to run them. Not everyone has the capital to purchase and maintain their own AI infrastructure, especially at the scale frontier-class models require.
That's where platforms like Vast.ai come in. By making GPU compute available on demand, we're lowering the barrier to entry for anyone who wants to run, train, fine-tune, or deploy open AI models, without the upfront costs. It's part of our mission to make AI accessible to everyone.
Whatever ends up happening with Hugging Face, the momentum behind open-source AI isn't slowing down -- and neither is the demand for compute.
Check out our Model Library to explore ready-to-deploy templates for popular open-source models like Kimi K3, Qwen3.8, Gemma 4, LTX 2.3, and many more. Then spin up a GPU instance and start building on Vast.ai today!
