Stable Point Aware 3D
Model Information
Display Name: Stable Point Aware 3D
API Model ID: stability-ai/stable-point-aware-3d
Category: Image To 3D
Description: Stable Point Aware 3D (SPAR3D) reconstructs the complete structure of a 3D object from a single image in seconds, with improved detail on unseen back regions. It combines point-cloud diffusion with mesh regression and allows editing of backside information. **Key Features:** - Single image to full 3D structure (GLB output) - Better depth and backside prediction than Stable Fast 3D - Point-cloud editing, remeshing and mesh simplification controls **Best For:** - Higher-fidelity 3D reconstruction from one image - Assets needing accurate back/unseen regions **Note:** Output is a binary GLB (model/gltf-binary). API in preview.
How to Use This Model
To use Stable Point Aware 3D via the HInow.ai API, use the model ID: stability-ai/stable-point-aware-3d
API Request Example (Image Generation)
POST https://api.hinow.ai/v1/images
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json
{
"model": "stability-ai/stable-point-aware-3d",
"prompt": "Your image description here"
}
Pricing
- total: $0.06
Available Parameters
- texture_resolution: Resolution of the albedo and normal map textures. (Options: 512, 1024, 2048)
- foreground_ratio: Padding around the object within the frame (1-2).
- remesh: Remeshing algorithm. quad/triangle useful for DCC tools. (Options: none, quad, triangle)
- target_type: Simplify to a target vertex or face count. (Options: none, face, vertex)
- target_count: Target vertex/face count (100-20000) when target_type is set. 1,000-10,000 recommended.
- guidance_scale: Guidance scaling of the point diffusion module (1-10). Default 3 gives best results.
- seed: Random seed for reproducibility. -1 (or omit) for a random seed.
Quick Reference
To use this model, set: "model": "stability-ai/stable-point-aware-3d"
Featured: No
Documentation: https://hinow.ai/models/stability-ai/stable-point-aware-3d
API Endpoint: https://api.hinow.ai/v1
