The reported Nvidia Hugging Face acquisition has moved beyond rumour and into the realm of near-certainty, with The Information citing a source familiar with the matter to report an agreed price of $12.9 billion. Business Insider, which broke news of the takeover interest at the weekend, added a note of caution: talks valuing the company at more than $13 billion had not yet produced a signed agreement and could still fall apart.
Neither Nvidia nor Hugging Face responded to requests for comment, though Nvidia’s silence carries its own weight: the company has a history of moving quickly to correct reports it considers wrong.
The Nvidia Hugging Face Acquisition Logic
For Nvidia, the strategic case is rooted in chip dependency. OpenAI, Google, Amazon, and Anthropic are all building proprietary silicon to reduce their reliance on Nvidia’s hardware. A sprawling open-source ecosystem, where developers pull models freely and run them on whatever compute is available, keeps more of the market anchored to Nvidia’s GPUs rather than to the closed labs’ custom chips.
That logic also explains Nvidia’s cloud ambitions. The company reportedly scaled back its DGX Cloud service roughly a year ago, but owning Hugging Face, which already lets developers run AI models on rented compute, would offer a re-entry into that market without rebuilding from scratch. There is a financial incentive, too: Nvidia has committed to cover the cost of tens of billions of dollars in cloud computing deals for customers. If that capacity goes unused, Hugging Face’s user base becomes a ready outlet for it.
Hugging Face CEO Clem Delangue has spent much of this year in public alignment with Nvidia’s open-source push. On CBS’s ‘Face the Nation’ this month, he said Hugging Face used an Nvidia-modified version of a Chinese open-source model to defend against a cyberattack, and he pointed to a letter, signed by Nvidia CEO Jensen Huang and 24 other companies including Hugging Face, urging the U.S. government to support rather than restrict open models. In a CNBC interview in late July, Delangue warned that China is ‘clearly dominating’ open-source AI.
From $4.5bn Platform to $12.9bn Prize
The price represents a dramatic re-rating. In August 2023, Hugging Face raised $235 million in a Series D round led by Salesforce Ventures, valuing the company at $4.5 billion, a figure that was already more than 100 times its annualised revenue at the time. The round brought Hugging Face’s total funding to $395.2 million, having received its first institutional cheque from Betaworks Ventures.
The investor list for that 2023 round reads like a roll-call of the companies now most anxious about Hugging Face’s independence. Beyond Salesforce Ventures, Reuters reported that participants included Google, Amazon, IBM, Intel, AMD, Qualcomm, Alphabet’s GV, Nvidia, and Sound Ventures, the fund backed by actor Ashton Kutcher. At the time of that round, Hugging Face employed 170 people.
The platform those 170 people had built was already substantial. Axios reported that as of August 2023, the site hosted 500,000 models, 250,000 datasets, and 250,000 apps, with the company aiming to triple those numbers through 2024. Hugging Face has described its ambition as building the GitHub for AI, a centralised hub where researchers, engineers, and data scientists converge.
The revenue picture has sharpened considerably since. The Information recently put Hugging Face’s annual revenue run-rate at roughly $150 million, up from about $100 million just two months earlier. Delangue told TechCrunch last month that the company was getting ‘close to profitability.’ A price near $13 billion would still represent a steep multiple, but rapid revenue growth compresses it.
This is not Nvidia’s first attempt. The Financial Times previously reported that Nvidia offered Hugging Face a $500 million investment at a $7 billion valuation late last year. Hugging Face declined, saying at the time it did not want a dominant investor capable of swaying its decisions. An outright acquisition changes that calculation: selling is categorically different from taking on a single large backer who holds leverage without full control.
The deal is taking shape against a wider consolidation wave in AI infrastructure. Stripe announced on 19 August 2026 that it had agreed to acquire OpenRouter, a model-routing startup founded in early 2023 that helps businesses select AI models based on their needs and budget. OpenRouter processes more than 10 trillion tokens daily from around 10 million developers and companies, charging a 5.5% fee on credit purchases, according to Payments Dive. The deal price was approximately $7.5 billion, per CNBC citing the New York Times, with $1.5 billion of that allocated to OpenRouter’s founders. The snippet described the price as ‘more than $7 billion’; the CNBC figure of approximately $7.5 billion is used here. OpenRouter had raised $113 million in its most recent funding round, less than three months before that announcement, at a $1.3 billion valuation.
The pattern is clear enough: the infrastructure layer of the AI economy is being absorbed rapidly, and at multiples that would have seemed extravagant even eighteen months ago. For Hugging Face, the question is no longer whether to sell but whether this particular buyer, at this particular price, is the one worth saying yes to. The next milestone is a signed agreement.
