Text-to-3D Workflow with Hyper3D
The text-to-3D workflow follows three steps once your API key is configured:1
Describe the asset
Tell Claude what you want. For example: “Generate a 3D model of a garden gnome.”
2
Claude calls the generation tool
Claude uses the Hyper3D integration to send your prompt to the Rodin API. The service generates a textured mesh.
3
Asset imports into Blender
The generated model is downloaded and placed in your active Blender scene, ready for positioning and rendering.
Image-to-3D Workflow
You can also generate assets from reference images. Provide a local file path or remote URL, and Claude sends it to Hyper3D or Hunyuan3D for generation. Example prompts:Combining Generated Assets with Library Assets
After generating a custom item, enhance it with library resources:- Apply Poly Haven textures to match the scene’s material palette
- Place the asset on Poly Haven surfaces (ground planes, floors)
- Set Poly Haven HDRI lighting to integrate the asset naturally into the environment
Post-Generation Refinement
Always verify generated assets before finalizing your scene:1
Check the bounding box
Use
get_object_info to see dimensions. Generated assets may be larger or smaller than expected.2
Adjust position, scale, and rotation
Ask Claude to center, scale, or rotate the asset to fit your scene.
3
Apply or replace materials
Use
execute_blender_code for custom material tweaks, or apply Poly Haven PBR textures for consistency.