Hyper3D Rodin Gen-2.5 is a text-and-image to 3D model from Deemos (Hyper3D) that turns one to five photos or a written prompt into a textured GLB asset. It targets game studios, e-commerce teams, and 3D artists who need production-ready meshes without hand modeling.
Core Features
- Up to five input images for image-to-3D, or pure text-to-3D, with an optional prompt to guide the reconstruction.
- Quality tiers from Minimum to Extreme-High, plus a target polygon budget, so you can iterate fast or push for final assets.
- Native 12K texture generation with PBR materials and a delit option for relightable surfaces that behave in any engine.
- Rig-ready T-pose or A-pose output and smart part-splitting that separates a model into printable pieces.
- Exports GLB, FBX, OBJ, USDZ, and STL, dropping straight into Blender, Maya, Unity, or Unreal.
Best For
- Game studios generating characters and hard-surface props for Unity or Unreal pipelines.
- E-commerce teams turning product photos into interactive 3D viewers for web or AR.
- Makers preparing models for 3D printing, where part-splitting enables multi-color output.
Pricing
Rodin runs on a credit system: new users get free starter credits, and paid credit packs unlock higher tiers and commercial export at the Extreme-High setting. Redo is free until you download, so iteration costs nothing until you commit. Enterprise and API access are available through Hyper3D for high-volume production.
Our Take
Best for teams that need detailed, rig-ready 3D without a modeling department. The trade-off is generation time at the top tier: Extreme-High renders around 80 seconds for up to 10 million polygons, so use the 4-second draft mode for early exploration. See more Image Generation tools.
