Sanja Fidler, VP of AI Research at NVIDIA, belongs to the AI Labs team in Toronto, which released the LATTE3D model and demonstrated it on-site at the GTC2024 spring session. The technology is capable of generating 3D models from text prompts with high efficiency, and can generate 3D models almost instantly (less than 1 second) when paired with NVIDIA's RTX A6000 GPU. With the NVIDIA RTX A6000 GPU, the 3D model can be generated almost instantly (less than 1 second), and the 3D shape can be enlarged and rotated at will, with no dead angle in 360 degrees. "
A year ago, it would have taken an hour for an AI model to generate a 3D visualization of this quality, whereas the current state of the art is about 10 to 12 seconds." Sanja Fidler said.
