Nvidia releases new version of open source AI model

Nvidia (NVDA.O) has released a series of new open-source AI models, stating that these models will be faster, cheaper, and smarter than its previous products. Nvidia is primarily known for providing chips, which companies like OpenAI use to train and profit from their closed-source models. However, Nvidia also offers a large number of its own models, covering fields ranging from physics simulations to autonomous vehicles. These models are provided as open-source software to researchers and other companies; for example, companies like Palantir Technologies have integrated Nvidia's models into their products. On Monday, Nvidia announced the third generation of its "Nemotron" large language model, primarily targeting tasks such as writing and programming. The smallest model, Nemotron 3 Nano, was released that day, with two larger versions expected in the first half of 2026. Meanwhile, reports indicate that Meta Platforms (META.O) is considering a shift to closed-source models, making Nvidia one of the leading open-source model providers in the US. (Jinshi)