Open-weight AI companies are attracting multibillion-dollar deals as larger technology firms seek control over model distribution, developer communities and inference demand. Nvidia is reportedly considering a $13 billion acquisition of Hugging Face, although neither company has confirmed an agreement.
The talks follow Nvidia’s $6 billion arrangement with open-weight model developer Poolside, under which most employees will move to the chipmaker, and Stripe’s purchase of model router OpenRouter for more than $7 billion. For Nvidia, owning a popular model platform could diversify its relationships beyond major cloud providers and frontier labs that are developing competing chips. For Stripe, routing many models can make token spending and payment infrastructure part of the same business.
Open-weight adoption remains limited but useful in repeatable, high-volume workloads. TechCrunch cites estimates that 6% of companies and 2% of software engineers currently use such models. Businesses often choose them for control and customization; self-hosting becomes more attractive once a workflow is stable enough to justify specialized engineering. The reported acquisitions do not mean open models are free to operate or automatically better than proprietary systems. They show that the surrounding platforms—hosting, routing, tuning tools and trusted developer access—can be more strategically valuable than the downloadable weights alone.