NVIDIA has released a new open-source AI model named Frontier, which is designed to rival GPT-4 and other large language models. The model is multimodal, featuring 530 billion parameters, and includes the training code. It performs closely to GPT-4o and Llama 3 70B in math and programming benchmarks, and supports vision tasks. This development marks a significant advancement in the AI field, positioning NVIDIA as a strong competitor in the market for advanced AI models. Additionally, NVIDIA has introduced a 72B model that supports similar benchmarks.