Browse ModelsTripo3dTripo3d H3.1 Multiview To 3d

Tripo3d H3.1 Multiview To 3d

Tripo3d H3.1 Multiview To 3d

Playground

Try it on WavespeedAI!

Tripo3D H3.1 Multiview-to-3D generates high-quality 3D models from 2-4 multi-angle images. Supports standard and HD texture quality, PBR materials, detailed geometry, quad mesh topology, texture alignment, and orientation control. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Tripo3D H3.1 Multiview-to-3D

Tripo3D H3.1 Multiview-to-3D reconstructs a production-ready 3D model from 2 to 4 reference images of the same object taken from different angles. Upload front, left, back, and right views — the model generates accurate geometry with optional PBR textures, quad mesh topology, and detailed quality tiers for both geometry and texture.


Why Choose This?

  • Multi-view 3D reconstruction Provide 2–4 images from different angles for significantly more accurate geometry than single-image reconstruction.

  • PBR material generation Generate Physically Based Rendering materials alongside the mesh for realistic real-time rendering in game engines and 3D tools.

  • Geometry and texture quality tiers Choose between standard and detailed quality independently for geometry and texture — optimizing for speed, cost, or maximum fidelity.

  • Quad mesh support Generate quad (4-sided) mesh topology instead of triangles for cleaner topology compatible with animation and sculpting workflows.

  • Texture alignment control Align textures to the original input image or to the generated geometry for different use case requirements.

  • Orientation control Use align_image to auto-rotate the model to match the input image orientation.


Parameters

ParameterRequiredDescription
imagesYes2–4 image URLs of the same object from different angles. Order: front, left, back, right. Front is required.
textureNoWhether to generate textures for the model. Default: true.
pbrNoWhether to generate PBR materials. Requires texture to be enabled. Default: true.
texture_qualityNoTexture quality: standard (default) or detailed (higher resolution).
geometry_qualityNoGeometry quality: standard (default) or detailed.
texture_alignmentNoHow textures are aligned: original_image (default) or geometry.
auto_sizeNoAuto-scale the model to real-world dimensions. Default: false.
orientationNoModel orientation: default or align_image (auto-rotates to match input image).
quadNoGenerate quad mesh topology instead of triangles. Default: false.

How to Use

  1. Upload your images — provide 2 to 4 image URLs of the same object from different angles. Front view is required; order as front, left, back, right.
  2. Configure texture settings (optional) — enable or disable texture and PBR generation, and choose standard or detailed texture quality.
  3. Set geometry quality (optional) — choose standard for faster results or detailed for higher geometric fidelity.
  4. Configure additional options (optional) — set texture alignment, orientation, quad mesh, and auto-size as needed.
  5. Submit — generate and download your 3D model.

Pricing

TextureTexture QualityGeometry QualityQuadCost
NoStandardNo$0.10
NoDetailedNo$0.30
YesStandardStandardNo$0.20
YesStandardDetailedNo$0.40
YesDetailedStandardNo$0.30
YesDetailedDetailedNo$0.50

Add +$0.05 to any combination above when quad is enabled.

Billing Rules

  • Base: $0.10 × texture multiplier (no texture: ×1, standard texture: ×2, detailed texture: ×3)
  • Detailed geometry surcharge: +$0.20
  • Quad surcharge: +$0.05
  • Default configuration (texture standard, geometry standard, no quad): $0.20

Best Use Cases

  • Game asset production — Reconstruct clean, textured 3D models from product or object photography for use in game engines.
  • E-commerce 3D — Generate 3D product models from multi-angle product photos for interactive viewers and AR.
  • Animation & VFX — Use quad mesh output for animation-ready topology compatible with rigging and sculpting workflows.
  • Digital twins — Accurately reconstruct real-world objects from multiple reference photographs.
  • AR/VR content — Generate PBR-textured models for immersive application development.

Pro Tips

  • Always include a clear front-facing image — it is required and the most important reference for accurate reconstruction.
  • More angles (3–4 images) produce significantly better geometry than using just 2.
  • Use detailed geometry quality for objects with fine surface detail like mechanical parts or organic shapes.
  • Use quad mesh when the model will be used for animation or further sculpting.
  • Enable auto_size when you need the model to match real-world scale.
  • Use align_image orientation when the input photo perspective matters for the final model alignment.

Notes

  • images is the only required field; all other parameters are optional.
  • Images should be ordered as: front, left, back, right. Front view is mandatory.
  • Please ensure your content complies with Tripo3D’s usage policies.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result


# Submit the task
curl --location --request POST "https://api.wavespeed.ai/api/v3/tripo3d/h3.1/image-to-3d" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
    "texture": true,
    "pbr": true,
    "texture_quality": "standard",
    "geometry_quality": "standard",
    "texture_alignment": "original_image",
    "auto_size": false,
    "orientation": "default",
    "quad": false
}'

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

Parameters

Task Submission Parameters

Request Parameters

ParameterTypeRequiredDefaultRangeDescription
imagesarrayYes[]2 ~ 4 items2 to 4 image URLs of the same object from different angles. Order: [front, left, back, right]. Front view is required.
texturebooleanNotrue-Whether to generate textures for the model.
pbrbooleanNotrue-Whether to generate PBR (Physically Based Rendering) materials. If true, texture is also enabled.
texture_qualitystringNostandardstandard, detailedQuality level for textures. 'detailed' produces higher-resolution textures.
geometry_qualitystringNostandardstandard, detailedQuality level for geometry.
texture_alignmentstringNooriginal_imageoriginal_image, geometryHow textures are aligned. 'original_image' aligns to input image, 'geometry' aligns to generated geometry.
auto_sizebooleanNofalse-Auto-scale the model to real-world dimensions.
orientationstringNodefaultdefault, align_imageModel orientation. 'align_image' auto-rotates to match the input image.
quadbooleanNofalse-Generate quad (4-sided) mesh topology instead of triangles.
face_limitintegerNo-1000 ~ 2000000Target number of faces for the generated mesh. If not set, the model adaptively determines the count.
model_seedintegerNo--Seed for geometry generation reproducibility.
texture_seedintegerNo--Seed for texture generation reproducibility.

Response Parameters

ParameterTypeDescription
codeintegerHTTP status code (e.g., 200 for success)
messagestringStatus message (e.g., “success”)
data.idstringUnique identifier for the prediction, Task Id
data.modelstringModel ID used for the prediction
data.outputsarrayArray of URLs to the generated content (empty when status is not completed)
data.urlsobjectObject containing related API endpoints
data.urls.getstringURL to retrieve the prediction result
data.has_nsfw_contentsarrayArray of boolean values indicating NSFW detection for each output
data.statusstringStatus of the task: created, processing, completed, or failed
data.created_atstringISO timestamp of when the request was created (e.g., “2023-04-01T12:34:56.789Z”)
data.errorstringError message (empty if no error occurred)
data.timingsobjectObject containing timing details
data.timings.inferenceintegerInference time in milliseconds

Result Request Parameters

ParameterTypeRequiredDefaultDescription
idstringYes-Task ID

Result Response Parameters

ParameterTypeDescription
codeintegerHTTP status code (e.g., 200 for success)
messagestringStatus message (e.g., “success”)
dataobjectThe prediction data object containing all details
data.idstringUnique identifier for the prediction, the ID of the prediction to get
data.modelstringModel ID used for the prediction
data.outputsstringArray of URLs to the generated content (empty when status is not completed).
data.urlsobjectObject containing related API endpoints
data.urls.getstringURL to retrieve the prediction result
data.statusstringStatus of the task: created, processing, completed, or failed
data.created_atstringISO timestamp of when the request was created (e.g., “2023-04-01T12:34:56.789Z”)
data.errorstringError message (empty if no error occurred)
data.timingsobjectObject containing timing details
data.timings.inferenceintegerInference time in milliseconds
© 2025 WaveSpeedAI. All rights reserved.