Create production-ready 3D models from text or images (up to 5) with rigging support
Fill in the parameters below and click "Generate" to try this model
Text prompt for 3D generation (optional for image mode - AI auto-generates if empty)
Input images (up to 5) for image-to-3D mode
Quality and mesh type selection
Output file format
Material type
Preserve transparency from input images
Generate in T-pose or A-pose (for character rigging)
Generate preview render image
Your inputs will be saved and ready after sign in
Professional image-to-3D with Hunyuan 3D Pro. Supports 8 views (front + optional back/left/right/top/bottom/left-front/right-front). Face count 40k-1.5M polygons
Turn sketches and line art into detailed 3D models. Combine sketch structure with text description for color, material, and category. Image resolution: 128x128 to 5000x5000 pixels
Reconstruct full 3D scenes with people and objects aligned in unified space.
Split 3D models into separate parts using Hunyuan 3D. FBX format only, max 100MB, ≤30k faces. Recommended for AIGC-generated models
Turn a single photo into accurate 3D body models with pose and shape.
Professional text-to-3D generation with Hunyuan 3D Pro. Advanced quality with adjustable face count (40k-1.5M polygons) and optional PBR materials (max 1024 characters)
Turn 1-4 images into production-ready 3D models with PBR textures
Production-ready 3D model from image. Quad/triangle topology, 100-300k polycount, PBR textures. Supports JPG, PNG, AVIF, HEIF. Outputs: GLB, FBX, USDZ, OBJ
Generate 3D human motions via text prompts using Hunyuan Motion 0.46B. Fast generation of realistic human animations in FBX format for 3D applications
Transforms text prompts and up to five reference images into high-quality, production-ready 3D models.
Offers multiple mesh qualities, from 2K for fast prototyping to 500K triangles for detailed, high-fidelity assets.
Supports a wide range of output formats including GLB, USDZ, FBX, OBJ, and STL for seamless workflow integration.
Provides both PBR and shaded material options for realistic or performance-optimized rendering.
Enables T-pose and A-pose model generation for easy character rigging and animation.
Allows transparency preservation from input images and generates preview render images for quick assessment.
Includes customization options like bounding box dimensions and random seed for reproducible, tailored results.
Rapid prototyping and asset creation for video games and interactive media.
Designing and visualizing characters, props, or products for animation and film production.
Generating optimized 3D models for AR and VR experiences.
Producing custom 3D assets for marketing, advertising, and e-commerce campaigns.
Enriching educational projects and presentations with unique 3D content.
Facilitating concept art development with AI-generated 3D references.
Supporting 3D printing workflows with exportable models in STL or OBJ formats.
3D artists, game developers, animators, marketers, AR/VR creators, and content designers seeking fast, high-quality 3D asset generation.
Start by entering a descriptive text prompt or upload up to five reference images (or use both for best results).
Select your desired mesh quality and type, choosing from quad or triangle options based on your workflow.
Pick the output file format that matches your preferred 3D software or platform (GLB, USDZ, FBX, OBJ, or STL).
Choose your material type—PBR, shaded, or both—and enable transparency preservation if needed.
For character models, enable T-pose or A-pose generation for easy rigging and animation.
Click to generate your 3D model and download the result once processing is complete; preview render images can be enabled for quick visual feedback.
You can generate a wide range of 3D models including characters, props, products, and conceptual objects from text prompts or reference images. The model supports both realistic and stylized outputs based on your input.
Hyper3D Rodin v2 supports GLB, USDZ, FBX, OBJ, and STL formats, making it compatible with most major 3D software and platforms. You can select your preferred format during generation.
Yes, you can upload up to five reference images for each generation. Combining images with a text prompt can improve the accuracy and detail of the generated 3D model.
Pricing varies by model and is based on a pay-as-you-go credit system. This lets you pay only for the assets you create, offering flexibility for different project needs.
Rodin v2 can generate models in T-pose or A-pose, which are ideal for character rigging and animation workflows. This makes it easy to integrate generated models into animation pipelines.
Hey! Need help? 👋
Click to chat with us