Hunyuan3d V2 Multi View

Hunyuan3d V2 Multi View

Playground

Hunyuan3D-V2-Multi-View is a state-of-the-art image-to-3D generative model developed by Tencent and now available on WaveSpeedAI.

Features

Hunyuan3D-V2-Multi-View

Hunyuan3D-2 is an open-source 3D generation model series launched by Tencent. As of March 2025, the latest version is 2.0, supporting the generation of high-fidelity 3D models with high-resolution texture maps through text, image, or sketch inputs.

Technical Highlights

The system adopts a separated process of geometry generation + texture synthesis:

-Geometry Generation (Hunyuan3D-DiT): Based on a flow diffusion model that generates untextured 3D geometric models, with 2.6B parameters, capable of precisely extracting geometric information from input images or text.

• Texture Synthesis (Hunyuan3D-Paint): Adds high-resolution (4K) textures to geometric models, with 1.3B parameters, supporting multi-view diffusion generation technology to ensure realistic textures and consistent lighting.
• By decoupling shape and texture generation, it effectively reduces complexity and improves generation quality.

Performance and Efficiency Optimization

• Fast Generation: Completes model generation in as fast as 30 seconds, with the accelerated version (Hunyuan3D-DiT-v2-0-Fast) shortening inference time by 50% through guidance distillation techniques.
• Multi-modal Input: Supports various input methods including text descriptions, images, and sketches, compatible with Blender plugins and Gradio applications, lowering the usage threshold.
• Open Source Model Ecosystem The project has open-sourced 6 models (some simplified versions), covering different scenario needs:

# Submit the task
curl --location --request POST "https://api.wavespeed.ai//api/v3/wavespeed-ai/hunyuan3d-v2-multi-view" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
    "front_image_url": "https://d3gnftk2yhz9lr.wavespeed.ai/media/images/1744448061406038945_mD5ACGR3.jpg",
    "back_image_url": "https://d3gnftk2yhz9lr.wavespeed.ai/media/images/1744448063989972538_psGEBxuq.jpg",
    "left_image_url": "https://d3gnftk2yhz9lr.wavespeed.ai/media/images/1744448066831654669_1hCzuqnj.jpg",
    "num_inference_steps": 50,
    "guidance_scale": 7.5,
    "octree_resolution": 256,
    "textured_mesh": false
}'

# Get the result
curl --location --request GET "https://api.wavespeed.ai//api/v3/predictions/${requestId}/result" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}"