Nvidia has officially agreed to acquire the open-source artificial intelligence (AI) platform Hugging Face for $12.9 billion. This deal signifies the AI chip giant’s gradual expansion beyond the hardware field into the higher reaches of AI technology stack.
This transaction accelerates Nvidia’s efforts to promote open-source AI models and services. These models and services aim to challenge the proprietary models of large AI companies in the United States and open-source models developed in China. The deal will further solidify Nvidia’s core position in the open-source model market. Open-source models refer to AI models that users can freely download, run, and customize, in contrast to the closed systems established by OpenAI and Anthropic.
With its widely-used Nemotron model, Nvidia has played an active role in supporting the open-source AI industry. Acquiring Hugging Face will allow direct access to a platform where developers can collaborate, test, and share tools, potentially gaining valuable insights and data to help narrow the technical gap with top US and Chinese labs.
Nvidia CEO Jensen Huang stated, “Hugging Face will continue to serve as an open platform that benefits the entire AI ecosystem.” He added that developers using Hugging Face for development or deployment will not be required to use Nvidia chips.
“We will jointly expand the scale of the Hugging Face platform, strengthen its infrastructure, and enable developers and institutions worldwide to access AI more widely,” Huang said in a statement on Thursday.
“Open-source models enable startups, enterprises, universities, and public institutions to develop on advanced AI capabilities without having to train each model from scratch,” he said.
Hugging Face, headquartered in New York and founded in 2016, has garnered support from Intel, AMD, and Amazon, and has emerged as one of the leading platforms to access open-weight AI models.
Under the agreement, Nvidia will pay approximately $11.9 billion to Hugging Face’s investors, while providing up to $1 billion in equity-based employee retention incentives for those joining Nvidia.
Nvidia’s introduction of the open-source AI model “Nemotron 3.5 Lightning” on August 11 garnered attention. According to Nvidia, this is an open 30-billion-parameter hybrid expert model designed for continuously running AI agents, enabling faster completion of large-scale, highly specialized tasks with output speeds up to four times faster than the equivalent scale model.