NVIDIA releases new open-source AI model

Odaily Planet Daily News: NVIDIA (NVDA.O) announced a series of new open-source artificial intelligence models, stating that these models will be faster, cheaper, and smarter than their previous products. NVIDIA is primarily known for providing chips, which companies like OpenAI use to train their closed-source models for profit. However, NVIDIA also offers a large number of proprietary models spanning various fields from physical simulation to autonomous vehicles, made available to researchers and other companies in the form of open-source software. For example, companies like Palantir Technologies have integrated NVIDIA’s models into their products. On Monday, NVIDIA unveiled the third-generation “Nemotron” large language model, mainly aimed at tasks like writing and programming. The smallest model, Nemotron 3 Nano, was released on the same day, with two larger versions scheduled for release in the first half of 2026. Meanwhile, reports indicate that Meta Platforms (META.O) is considering shifting to closed-source models, making NVIDIA one of the leading open-source model providers in the United States. (Jin10)