NVIDIA has finally, after weeks of rumors, confirmed the acquisition of the AI platform Hugging Face for $12.93 billion; a deal that could strengthen NVIDIA’s position in software and the AI model ecosystem.
The Hugging Face platform currently hosts about 3 million models, one million applications, and 500,000 datasets, and more than 18 million developers use its services.
Hugging Face will remain open source
Jensen Huang, NVIDIA’s CEO, said in a blog post that after the deal, Hugging Face will continue to support open-source models and will keep its focus on increasing developers’ access to AI tools.
Huang emphasized that Hugging Face will operate as an open platform for the entire AI ecosystem. According to him, developers will still be able to choose the models, frameworks, cloud services, inference providers, and compute infrastructure they prefer, and using NVIDIA’s compute infrastructure through Hugging Face will not be mandatory.
The NVIDIA CEO also referred to the company’s previous collaborations with Hugging Face and said NVIDIA has so far published more than 500 models and 250 open datasets on the platform.
Why is acquiring Hugging Face important for NVIDIA?
NVIDIA is currently the main player in the hardware market used to train and run AI models. Acquiring Hugging Face could enable the company to expand its presence in AI software and services as well.
Hugging Face hosts one of the most important developer communities for open-source models, and controlling this ecosystem could help NVIDIA build a broader platform around its hardware.
On the other hand, the deal could enable better utilization of NVIDIA’s compute capacity. The company could make part of the unused capacity of its data centers available to enterprise customers who use Hugging Face’s services.
The growth history of Hugging Face
Hugging Face was founded in 2016 and, according to Crunchbase data, has raised more than $395 million to date. Its most recent funding round took place in 2023, when Hugging Face secured $235 million from Salesforce Ventures and other investors, including Google, Amazon, IBM, and NVIDIA.
Clément Delangue, CEO of Hugging Face, also thanked the platform’s user community in a post on X, saying the community has shown that Hugging Face can act as an alternative to open AI APIs.
Delangue believes that to achieve this goal at a much larger scale, Hugging Face needs more compute capacity, support, collaboration, and transparency. According to him, Jensen Huang’s offer was precisely focused on providing these capacities.
Hugging Face’s $150 million revenue
Over recent years, by continuously increasing the number of models offered on its platform, Hugging Face has become one of the recognized names in the AI industry.
According to published reports, the company rejected NVIDIA’s $500 million offer last year. Also, The Information reported last month that Hugging Face’s annual revenue has reached about $150 million.
Delangue also said in an interview in July 2026 that the platform’s rapid growth has brought Hugging Face close to profitability.
NVIDIA’s extensive investment in AI models
The acquisition of Hugging Face is part of NVIDIA’s broader strategy to have a stronger presence in the field of AI models. Huang has repeatedly supported open-source models and even signed a letter with several other organizations emphasizing the importance of open models to bolster the United States’ position in the AI competition with countries like China.
NVIDIA is also heavily investing in developing AI models. According to the Wall Street Journal, last month the company signed a $6 billion contract with coding startup Poolside to develop open-source models.
In its recent financial report, NVIDIA also stated that it has provided more than $50 billion to leading AI labs.
In response to an analyst’s question about open-source models, Huang stressed their importance for the future of the AI industry and said that almost all open-source models run on NVIDIA hardware.
Open-source models and cybersecurity
The NVIDIA CEO also considered open-source models important for areas such as cybersecurity. According to him, advanced AI models allow security companies to create autonomous and large-scale defense systems, and open-source models play an important role in providing such capabilities.
This matter gained more significance when Clément Delangue said in July 2026 that one of NVIDIA’s open-source models helped Hugging Face protect its infrastructure against cyberattacks; while proprietary models had failed to defend the platform effectively in that situation.
Days earlier, OpenAI had also confirmed that an unreleased model from the company had been exfiltrated to Hugging Face.
The $12.93 billion acquisition of Hugging Face ultimately shows that NVIDIA is no longer solely seeking dominance over the AI hardware layer and is aiming to expand its presence in software, models, and the developer community as well. If Hugging Face maintains its open-source approach, the deal could be one of NVIDIA’s most important steps toward creating a comprehensive ecosystem in the AI industry.