Key takeaways
- Nvidia interim CEO Collette Kress said the company remains committed to open-source AI despite founder Jensen Huang's departure over personal conduct and alleged Chinese spy connections.
- Weekly token volume on OpenRouter surged 35-fold year-over-year, with Nvidia's Nemotron model ranking as the most widely used U.S. open model on the platform.
- Nvidia's $12.9 billion acquisition of Hugging Face, which serves 118 million developers and 200,000 companies, remains on track despite the leadership transition.
Nvidia interim CEO Collette Kress used Tuesday’s earnings call to reaffirm the company’s commitment to open-source artificial intelligence, the first time company leadership has publicly addressed strategy since founder Jensen Huang’s resignation. “At Nvidia, we are committed to fostering a vibrant open model ecosystem, so every country, company, and industry can build its own intelligence,” Kress said in her opening remarks.
Her statement directly addressed concerns raised by Huang’s departure. The founder’s resignation, announced after revelations of extramarital affairs including one with Alice Huang, an executive at Singapore-based Megaspeed who was allegedly a Chinese spy, had prompted questions about whether Nvidia might retreat from its support for non-proprietary AI development. Hugging Face CEO responded to the news by calling it “a sad day” and pledging to “keep fighting for open-source AI, harder than ever,” underscoring the stakes for the broader developer community.
Nvidia faces additional pressure beyond the leadership transition. The SEC is investigating the company over whether its public disclosures adequately informed investors, according to The Wall Street Journal. Huang is set to testify before Congress by Thanksgiving regarding his personal conduct.
Market Data Shows Accelerating Open Model Adoption
Kress grounded her reaffirmation of commitment in concrete market figures. She cited statistics from OpenRouter showing weekly token volume surged 35-fold year-over-year, with an additional 3.5-fold increase over the past three months. This represents not theoretical interest but actual usage patterns—developers building real AI systems on open models rather than relying exclusively on proprietary alternatives.
Nemotron Leads U.S. Open Model Deployment
Nvidia’s own open-source models are capturing meaningful share of this growth. The Nemotron family of open-weight models, developed under Huang’s direction, ranks as the most widely used U.S. open model on OpenRouter. Nemotron was built for long-running agentic AI systems and reasoning workloads, filling specific use cases that developers want to implement without dependency on closed systems.
Kress described Nemotron as “where developers start to build their own intelligence.” This positioning matters strategically for Nvidia. Rather than creating bottlenecks that force adoption of proprietary models, the company benefits when developers choose its hardware platforms as the foundation for whatever models they want to run, whether open or closed.
Hugging Face Acquisition Continues on Schedule
Nvidia announced its acquisition of Hugging Face for $12.9 billion in September—the company’s second-largest deal in history. Despite the leadership upheaval, the acquisition remains on track.
Scale of the Hugging Face Platform
Hugging Face operates the most widely used platform for discovering and sharing open-source AI models. It serves 118 million developers and 200,000 companies globally. For teams outside Nvidia that want to build custom AI systems, Hugging Face provides critical infrastructure—model discovery, training resources, and community features that have become central to open-source AI development.
What Nvidia’s Ownership Could Mean
The acquisition itself raised questions within the developer community about independence and control. Would a Nvidia-owned Hugging Face face pressure toward proprietary interests? Would the platform’s commitment to openness remain intact under corporate ownership?
Kress directly addressed these concerns, albeit briefly. “Together, we will scale its platform and make intelligence more open, more capable, and more accessible,” she said, suggesting Nvidia sees opportunity in expanding rather than restricting Hugging Face’s scope. The acquisition was not discussed further during the earnings call’s question-and-answer segment, leaving details about timeline and integration plans for future disclosure.

Huang’s Personal Advocacy and Its Absence
Jensen Huang’s role in advocating for open-source AI within Nvidia cannot be separated from his personal prominence and influence. He had pushed the company to develop Nemotron and to support open-model ecosystems even when proprietary alternatives might generate more direct revenue.
His departure created a credibility gap for open-source advocates. Would a new leadership team maintain that commitment without Huang’s visible championing? Would Nvidia gradually shift resources toward proprietary systems? These questions were not purely speculative—corporate strategy does change with leadership transitions.
Huang’s exit also arrived with significant complications. Beyond the affairs involving two direct reports, his alleged involvement with Alice Huang at Megaspeed introduced additional complexity around both personal conduct and potential national security concerns. These dimensions made his departure a matter of broader industry attention beyond typical executive transitions.
What Market Momentum Suggests
The token volume growth that Kress cited reflects genuine market momentum independent of Nvidia’s leadership. Developers are building on open models because they provide value and flexibility, not primarily because Nvidia endorses them. This suggests that even if Nvidia’s support ebbed, the open-source AI ecosystem possesses its own momentum.
Still, Nvidia’s backing remains material. The company manufactures the chips developers use to train and run AI models. Its commitment to treating open and proprietary systems as complementary rather than competing provides assurance that hardware optimization will continue across the landscape.
The immediate question is whether the Hugging Face acquisition will close as anticipated and on what timeline. Beyond that, how Nvidia’s ownership might eventually reshape the platform—its governance, its incentives, its relationship to commercial interests—will determine whether this represents genuine alignment with open-source values or a longer-term consolidation of control.
Source: TechCrunch
Frequently Asked Questions
Why did Jensen Huang resign from Nvidia?
Huang resigned following revelations of extramarital affairs, including one with Alice Huang, an executive at Singapore-based Megaspeed who was allegedly a Chinese spy, and two affairs with direct reports at Nvidia. He is scheduled to testify before Congress by Thanksgiving.
What is Nemotron, and why does it matter?
Nemotron is Nvidia's family of open-weight AI models designed for long-running agentic systems and reasoning workloads. It currently ranks as the most widely used U.S. open model on OpenRouter, suggesting strong developer adoption of Nvidia's open-source AI alternatives.
Is Nvidia still acquiring Hugging Face?
Yes. Interim CEO Collette Kress confirmed that the $12.9 billion acquisition of Hugging Face, announced in September, remains on track despite Huang's resignation and continues as a strategic priority for the company.