tencent/hunyuan3d-3.1

Generate high-quality 3D models with accurate geometry and realistic textures from input images.

$0.16 / request
GPU: A100

Pricing

tencent/

hunyuan3d-3.1

Pricing for Synexa AI models works differently from other providers. Instead of being billed by time, you are billed by input and output, making pricing more predictable.

Output
$0.1600 / 3d
or
6 3ds / $1

For example, generating 100 3ds should cost around $16.00.

Check out our docs for more information about how per-request pricing works on Synexa.

ProviderPrice ($)Saving (%)
Synexa$0.1600-
replicate$0.16000.0%

Readme

Hunyuan3D 3.1 is part of Tencent's Hunyuan3D family of generative AI models designed to create high-quality 3D assets from images.

Features

  • Image-to-3D Conversion: Convert reference images into textured 3D models
  • Improved Geometry Quality: Enhanced mesh topology and surface consistency
  • High-Resolution Texturing: Produces realistic materials and lighting effects

Use Cases

  • Game development
  • Virtual and augmented reality
  • Animation and film pre-visualization
  • Product visualization and e-commerce
  • 3D printing and rapid prototyping

Architecture

  1. Image encoder converts input into latent representations
  2. Diffusion models generate 3D geometry
  3. Geometry converted into usable mesh formats (GLB)
  4. Textures synthesized and mapped to the mesh

Recommendations

  • Use images with simple backgrounds, no text, single object
  • Object should occupy >50% of the frame
  • Supported formats: jpg, png, jpeg, webp
  • Resolution: 128-5000px per side, max 6MB