NVIDIA's GR00T-N1.7-AppleToPlate Model Release
NVIDIA releases the GR00T-N1.7-AppleToPlate model with 3 billion parameters.
NVIDIA has released the GR00T-N1.7-AppleToPlate model, featuring 3 billion parameters and utilizing F32 tensor types. This model is designed for multimodal applications, though specific details on its capabilities, context window, or pricing are not provided. Currently, the model is not deployed by any inference provider, and interested parties are encouraged to request provider support for deployment. According to the available information on Hugging Face, the model’s weights are accessible in safetensors format, but further claims about its performance or advantages over previous models are not explicitly stated by NVIDIA.