Several Nvidia researchers generously contributed their computing resources to expedite the launch of Nvidia’s Llama Nemotron models. According to Jonathan Cohen, Vice President of Applied Research at Nvidia, the rapid development of these models was made possible by researchers within the company being willing to share their computational power. In an interview, Cohen highlighted the importance of GPU access as a key currency in the field of AI research. The development speed of AI systems is heavily reliant on available computing resources, as explained by Cohen.
Under Cohen’s leadership, Nvidia successfully introduced the Llama Nemotron models into the market, marking the company’s venture into reasoning AI systems. The efficient development process, taking only one to two months, was partly due to the altruistic contributions of researchers who sacrificed their computing power. Cohen praised Nvidia’s culture of prioritizing vital projects and the collaborative effort across diverse teams to achieve this milestone.
The development of Llama Nemotron demanded sacrifices in terms of computational resources and personnel, but individuals set aside personal interests for the collective good. Cohen commended the leadership and selfless decisions that culminated in the successful creation of the models. Nvidia did not provide a comment to Business Insider at the time of publication.
