Nvidia 3D Gen

The Nvidia 3D Gen skill enables AI agents to create 3D assets from text prompts or 2D images. It interfaces with the NVIDIA TRELLIS API, which is hosted on NVIDIA NIM, to generate production-ready 3D meshes. This capability offers two primary functions: text-to-3D, converting descriptive textual prompts into textured 3D meshes, and image-to-3D, transforming 2D reference images and concept sketches into their 3D volumetric equivalents. The skill manages underlying data operations such as base64 encoding and decoding, asynchronous request polling, and state management required for multi-stage generation processes. The output is provided as standard GLB files. These files are immediately ready for use in professional pipelines and can be directly imported into common 3D development environments including Blender, Unity, Unreal Engine, Godot, and Three.js. This allows an agent to seamlessly integrate generated assets into game development, simulations, or web-based 3D applications. The skill also includes automatic local directory management with timestamped versioning for organized storage of generated 3D models.