AI & text-to-CAD
Text-to-3D
Definition
Text-to-3D is AI that generates a 3D shape from a text prompt for visuals, games, or 3D printing, usually as a mesh or neural representation. It targets appearance rather than exact dimensions or editable CAD features.
Also known as: text to 3D, AI 3D model generator
Updated
Where it comes from#
Text-to-3D research grew out of image generation. DreamFusion used a 2D diffusion model to optimize a 3D scene from text, and Shap-E generates implicit 3D functions that can be rendered or turned into meshes.
The output is usually exported as OBJ, glTF, or STL. It can look convincing and print well as a figurine, but it has no exact faces, dimensions, or feature history. For parts that need to fit, be machined, or be edited, text-to-CAD is the relevant category.
Sources
Related terms
Text-to-CAD
Text-to-CAD is AI that turns a natural-language description into editable CAD geometry, such as a STEP file, a parametric model, or CAD code. The goal is a model with exact faces and dimensions that can be modified and manufactured, not just a shape that looks right on screen.
Mesh
A mesh represents a shape as a surface of small flat polygons, almost always triangles, joined at shared edges and vertices. Meshes work well for rendering, games, and 3D printing, but they have no true faces, edges, or exact curves, so they are hard to dimension or edit as engineering geometry.
glTF
glTF is a Khronos Group format for delivering 3D scenes and models efficiently to web and real-time applications. It stores meshes with materials, textures, and animation, which makes it good for viewers and previews but not for engineering edits.