Skip to main content
ModelRunner
hyper3d avatar

hyper3d / rodin

Generate a production-ready, textured 3D model (GLB, USDZ, FBX, OBJ, or STL) from a text prompt or one or more reference images.

0.4
OpenAPI

Model Input

Input

Text description of the 3D object to generate. Required for text-to-3D (provide this OR one or more input images).

  • https://media.modelrunner.ai/AOPTjhPDbZLs8yy1Qtrht.png

One or more image URLs to reconstruct into a 3D model. Required for image-to-3D (provide these OR a text prompt). Multiple images are combined per the condition_mode setting.

Additional Settings

Customize your input with more control.

How multiple input images are combined: concat treats them as different views of the same object; fuse blends features from distinct references. Only relevant when input_image_urls has more than one image.

Random seed for reproducible generation. Leave unset for a random result.

Export format of the generated 3D mesh.

Material type baked into the mesh: PBR (physically based) or Shaded.

Generation quality; higher values yield denser geometry and richer textures at the cost of speed.

Enable enhanced geometry export for finer surface detail.

Generation tier: Regular for full generation, or Sketch for a faster rough pass.

Generate human-like models in a rig-friendly T/A pose.

You need to be logged in to run this model and view results.
Log in

Model Output

Output

Loading
Generated in 98.316 seconds
Logs (1 lines)

Model Example Requests

Examples

Model Details

Model Details

Rodin by Hyper3D turns a text description or one or more reference images into a complete, textured 3D model. You give it a prompt (text-to-3D), an image or set of images (image-to-3D), or both, and it returns a downloadable mesh with PBR materials baked in — geometry plus surface material, not just a point cloud. Choose the export format (GLB, USDZ, FBX, OBJ, or STL) and quality tier to fit your pipeline, from quick previews to high-detail production assets. Its standout is handling both input modes through a single endpoint and emitting genuinely production-ready assets suitable for games, AR/VR, and 3D printing.

## Best for - Turning a product or concept photo into a textured 3D asset for games or AR - Generating a 3D model from a text description when you have no reference image - Reconstructing an object from several views of the same subject for cleaner geometry - Producing meshes in a specific format your engine needs (USDZ for Apple AR, FBX/OBJ for DCC tools, STL for 3D printing) - Human or character models where a neutral T/A pose helps rigging

## Choose another model when - You only need a flat 2D image, not a 3D mesh — use a text-to-image or image-to-image model - You need an animated 3D scene or video rather than a static mesh — use an image-to-video model - You want to edit an existing photo rather than reconstruct geometry — use an image-editing model

## Tips - Provide either a `prompt` or `input_image_urls` (or both). Sending neither fails at generation time. - For multi-image input, set `condition_mode` to `concat` when the images are different views of the same object, or `fuse` when combining features from distinct references. - Set `geometry_file_format` to match your target pipeline; the output URL always points to a mesh in the chosen format. - Enable `TAPose` for human-like models to get a rig-friendly T/A pose.

## Advanced Configuration - `condition_mode`: `concat` (multi-view of one object, default) or `fuse` (blend features from distinct reference images). Only meaningful with multiple `input_image_urls`. - `quality`: `high`, `medium` (default), `low`, or `extra-low` — trade detail against speed. Higher quality yields denser geometry and richer textures. - `tier`: `Regular` (default) for full generation, or `Sketch` for a faster rough pass. - `material`: `PBR` (default, physically based) or `Shaded`. - `use_hyper`: enable enhanced geometry export for finer surface detail.

To run via the ModelRunner JavaScript client: ```js import { modelrunner } from "@modelrunner/client";

const result = await modelrunner.subscribe("hyper3d/rodin", { input: { prompt: "A simple wooden stool", geometry_file_format: "glb", quality: "medium", tier: "Regular", material: "PBR", }, }); ```