3D Generation Models - Parameters Reference

This page is auto-generated from model configurations. Last updated: 2026-03-13.

This reference lists all available 3d generation models and their parameters. Use these parameter names when calling the Generation API.

Models

Scenario

Scenario Voxels

Model ID: model_scenario-voxels

Capabilities: txt23d, img23d

LLM Markdown: https://app.scenario.com/api/models/model_scenario-voxels/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`prompt`	string	Yes	-	-	-	-	Text prompt for voxel generation
`referenceImages`	file_array	No	-	-	-	-	List of input images for voxel generation
`width`	number	No	`32`	16	256	-	Width of the voxel grid
`height`	number	No	`32`	16	256	-	Height of the voxel grid
`depth`	number	No	`32`	16	256	-	Depth of the voxel grid

Academia

PartCrafter

PartCrafter is a 3D model generator that creates 3D models from images.

Model ID: model_partcrafter

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_partcrafter/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`images`	file	Yes	-	-	-	-	Input image to generate 3D asset from
`numParts`	number	No	`6`	1	16	-	Number of parts to generate
`steps`	number	No	`75`	1	100	-	Number of inference steps.
`guidanceScale`	number	No	`7`	1	10	-	Higher values will keep output closer to the input image
`numTokens`	number	No	`4096`	1024	4096	-	Number of tokens
`seed`	number	No	-	0	2147483647	-	Use a seed for reproducible results. Leave blank to use a random seed.

Ultrashape 1.0

UltraShape-1.0 is a 3D diffusion framework that generates high-fidelity 3D geometry through coarse-to-fine geometric refinement.

Model ID: model_ultrashape-1-0

Capabilities: 3d23d

LLM Markdown: https://app.scenario.com/api/models/model_ultrashape-1-0/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`image`	file	Yes	-	-	-	-	Reference image for mesh refinement.
`model`	file	Yes	-	-	-	-	The coarse mesh to refine
`numInferenceSteps`	number	No	`50`	1	50	-	Number of inference steps.
`octreeResolution`	number	No	`1024`	128	1024	-	Marching cubes resolution.
`seed`	number	No	-	0	-	-	Seed value for randomization, leave blank to use a random seed

Deemos Technology

Rodin Hyper3D

Rodin by Hyper3D generates realistic and production ready 3D models from text or images.

Model ID: model_rodin-hyper3d

Capabilities: txt23d, img23d

LLM Markdown: https://app.scenario.com/api/models/model_rodin-hyper3d/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`prompt`	string	No	-	-	-	-	A textual prompt to guide model generation. Required for Text-to-3D mode. Optional for Image-to-3D mode.
`images`	file_array	No	-	-	-	-	Images to use while generating the 3D model. Required for Image-to-3D mode. Optional for Text-to-3D mode.
`conditionMode`	string	No	`concat`	-	-	`fuse`, `concat`	For fuse mode, One or more images are required. It will generate a model by extracting and fusing features of objects from multiple images. For concat mode, need to upload multiple multi-view images of the same object and generate the model.(You can upload multi-view images in any order, regardless of the order of view.) Default value: 'concat'
`material`	string	No	`PBR`	-	-	`PBR`, `Shaded`	Material type. Possible values: PBR, Shaded.
`quality`	string	No	`high`	-	-	`high`, `medium`, `low`, `extra-low`	Generation quality.
`useHyper`	boolean	No	`false`	-	-	-	Whether to export the model using hyper mode.
`tier`	string	No	`Regular`	-	-	`Sketch`, `Regular`	Tier of generation. For Rodin Sketch, set to Sketch. For Rodin Regular, set to Regular.
`TAPose`	boolean	No	`false`	-	-	-	When generating the human-like model, this parameter control the generation result to T/A Pose.
`seed`	number	No	-	0	65535	-	Seed value for randomization, leave blank to use a random seed

Rodin Hyper3D Gen-2

Rodin by Hyper3D generates realistic and production ready 3D models from text or images.

Model ID: model_rodin-hyper3d-v2

Capabilities: txt23d, img23d

LLM Markdown: https://app.scenario.com/api/models/model_rodin-hyper3d-v2/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`prompt`	string	No	-	-	-	-	A textual prompt to guide model generation. Required for Text-to-3D mode. Optional for Image-to-3D mode.
`images`	file_array	No	-	-	-	-	Images to use while generating the 3D model. Required for Image-to-3D mode. Optional for Text-to-3D mode.
`material`	string	No	`All`	-	-	`All`, `PBR`, `Shaded`	Material type. Possible values: PBR, Shaded.
`quality`	string	No	`18K Quad`	-	-	`4K Quad`, `8K Quad`, `18K Quad`, `50K Quad`, `2K Triangle`, `20K Triangle`, `50K Triangle`, `150K Triangle`, `500K Triangle`	Generation quality.
`TAPose`	boolean	No	`false`	-	-	-	When generating the human-like model, this parameter control the generation result to T/A Pose.
`useOriginalAlpha`	boolean	No	`false`	-	-	-	When enabled, preserves the transparency channel from input images during 3D generation.
`seed`	number	No	-	0	65535	-	Seed value for randomization, leave blank to use a random seed

Rodin Hyper3D HighPack

Rodin by Hyper3D generates realistic and production ready 3D models from text or images. Will provide 4K resolution textures instead of the default 1K, as well as models with high-poly.

Model ID: model_rodin-hyper3d-highpack

Capabilities: txt23d, img23d

LLM Markdown: https://app.scenario.com/api/models/model_rodin-hyper3d-highpack/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`prompt`	string	No	-	-	-	-	A textual prompt to guide model generation. Required for Text-to-3D mode. Optional for Image-to-3D mode.
`images`	file_array	No	-	-	-	-	Images to use while generating the 3D model. Required for Image-to-3D mode. Optional for Text-to-3D mode.
`conditionMode`	string	No	`concat`	-	-	`fuse`, `concat`	For fuse mode, One or more images are required. It will generate a model by extracting and fusing features of objects from multiple images. For concat mode, need to upload multiple multi-view images of the same object and generate the model.(You can upload multi-view images in any order, regardless of the order of view.) Default value: 'concat'
`material`	string	No	`PBR`	-	-	`PBR`, `Shaded`	Material type. Possible values: PBR, Shaded.
`quality`	string	No	`high`	-	-	`high`, `medium`, `low`, `extra-low`	Generation quality.
`useHyper`	boolean	No	`false`	-	-	-	Whether to export the model using hyper mode.
`tier`	string	No	`Regular`	-	-	`Sketch`, `Regular`	Tier of generation. For Rodin Sketch, set to Sketch. For Rodin Regular, set to Regular.
`TAPose`	boolean	No	`false`	-	-	-	When generating the human-like model, this parameter control the generation result to T/A Pose.
`seed`	number	No	-	0	65535	-	Seed value for randomization, leave blank to use a random seed

Hitem3D

Sparc3D

Model ID: model_sparc3d

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_sparc3d/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`images`	file_array	Yes	-	-	-	-	Upload 1-4 images, starting with the front view. For multi-image input, you can optionally include back, left, and right views (in that order).
`requestType`	number	No	`3`	-	-	`1`, `3`	Choose whether to generate just a mesh or a textured 3D asset
`resolution`	string	No	`1024`	-	-	`512`, `1024`, `1536`, `1536pro`	Controls the voxel resolution of the generated 3D asset. Higher values deliver more detail and cleaner topology.

Sparc3D 2.0

Model ID: model_hitem-sparc-3d-2

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_hitem-sparc-3d-2/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`images`	file_array	Yes	-	-	-	-	Upload 1-4 images, starting with the front view. For multi-image input, you can optionally include back, left, and right views (in that order).
`requestType`	number	No	`3`	-	-	`1`, `3`	Choose whether to generate just a mesh or a textured 3D asset
`resolution`	string	No	`1536`	-	-	`1536`, `1536pro`	Controls the voxel resolution of the generated 3D asset. Higher values deliver more detail and cleaner topology.

Sparc3D 2.0 Portrait

Model ID: model_hitem-sparc-3d-portrait-2

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_hitem-sparc-3d-portrait-2/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`images`	file_array	Yes	-	-	-	-	Upload 1-4 images, starting with the front view. For multi-image input, you can optionally include back, left, and right views (in that order).
`requestType`	number	No	`3`	-	-	`1`, `3`	Choose whether to generate just a mesh or a textured 3D asset

Sparc3D Portrait

Model ID: model_sparc3d-portrait

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_sparc3d-portrait/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`images`	file_array	Yes	-	-	-	-	Upload 1-4 images, starting with the front view. For multi-image input, you can optionally include back, left, and right views (in that order).
`requestType`	number	No	`3`	-	-	`1`, `3`	Choose whether to generate just a mesh or a textured 3D asset

Meshy

Meshy Image-to-3D

Model ID: model_meshy-img23d

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_meshy-img23d/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`aiModel`	string	No	`latest`	-	-	`meshy-4`, `meshy-5`, `latest`	Meshy model to use.
`modelType`	string	No	`standard`	-	-	`standard`, `lowpoly`	Specify the type of 3D mesh generation. 'standard' for regular high-detail 3D mesh generation, 'lowpoly' for low-poly mesh optimized for cleaner polygons. When lowpoly is selected, following parameters are ignored: model, topology, target polycount, should remesh.
`image`	file_array	Yes	-	-	-	-	Upload one image to convert it into 3D, or provide up to four images from multiple angles for higher accuracy.
`texturePrompt`	string	No	-	-	-	-	Provide a text description to guide the texturing process. The Style Prompt is only used when no Style Image is provided. If you upload an image, this text prompt will be ignored.
`textureImage`	file	No	-	-	-	-	Upload a reference image to guide the visual style of the generated texture. This will override the Style Prompt. Requires the Texture setting to be enabled (toggle ON).
`shouldTexture`	boolean	No	`true`	-	-	-	Enable texture generation for a fully colored model. Disable to create a plain, untextured mesh.
`topology`	string	No	`triangle`	-	-	`triangle`, `quad`	Choose the mesh structure type used for the generated model (e.g., triangles or quads).
`enablePbr`	boolean	No	`true`	-	-	-	Generate advanced PBR maps (metallic, roughness, normal) for more realistic materials. Requires the Texture setting to be enabled (toggle ON).
`poseMode`	string	No	-	-	-	``, `a-pose`, `t-pose`	Specify the pose mode for the generated model. Accepts 'A-Pose', 'T-Pose', or empty string (default).
`shouldRemesh`	boolean	No	`false`	-	-	-	Enable remeshing to improve mesh quality and control polycount. Disable to keep the raw, unprocessed mesh.
`symmetryMode`	string	No	`auto`	-	-	`off`, `on`, `auto`	Control how symmetry is applied during model generation for balanced, even shapes.
`targetPolycount`	number	No	`30000`	100	300000	-	Specify the target number of polygons in the generated model. The actual number of polygons may deviate from the target depending on the complexity of the geometry.

Meshy Remesh

Model ID: model_meshy-remesh

Capabilities: 3d23d

LLM Markdown: https://app.scenario.com/api/models/model_meshy-remesh/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`model`	file	Yes	-	-	-	-	Upload a GLB file of the 3D model you want to remesh.
`topology`	string	No	`triangle`	-	-	`triangle`, `quad`	Choose the mesh structure type used for the generated model. For Quad, choose output format 'obj'
`targetPolycount`	number	No	`30000`	100	300000	-	Set the approximate number of polygons for your 3D model. Higher values increase detail, while lower values create simpler meshes.
`resizeHeight`	number	No	`0`	0	10000	-	Adjust the model's height in meters. Set to 0 to keep the original scale.
`originAt`	string	No	`empty`	-	-	`empty`, `bottom`, `center`	Define where the model's origin point is positioned. Leave empty to keep the default origin.

Meshy Retexture

Retexture your 3D model with Meshy's AI re texturing capabilities.

Model ID: model_meshy-retexture

Capabilities: 3d23d

LLM Markdown: https://app.scenario.com/api/models/model_meshy-retexture/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`model`	file	Yes	-	-	-	-	Upload a GLB file of the 3D model you want to retexture.
`textStylePrompt`	string	No	-	-	-	-	Describe your desired texture style of the object using text.
`imageStyle`	file	No	-	-	-	-	Provide a 2d image to guide the texturing process.
`enablePbr`	boolean	No	`false`	-	-	-	Generate PBR Maps (metallic, roughness, normal) in addition to the base color.
`enableOriginalUv`	boolean	No	`true`	-	-	-	Use the model's existing UV map instead of generating a new one. If no UVs are present, output quality may be lower.

Meshy Rigging

Model ID: model_meshy-rigging

Capabilities: 3d23d

LLM Markdown: https://app.scenario.com/api/models/model_meshy-rigging/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`model`	file	Yes	-	-	-	-	The 3D model (GLB)
`heightMeters`	number	No	`1.7`	0.1	100	-	The approximate height of the character model in meters. This aids in scaling and rigging accuracy. It must be a positive number.
`textureImage`	file	No	-	-	-	-	Model's base color texture image.

Meshy Text-to-3D

Model ID: model_meshy-txt23d

Capabilities: txt23d

LLM Markdown: https://app.scenario.com/api/models/model_meshy-txt23d/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`aiModel`	string	No	`latest`	-	-	`meshy-4`, `meshy-5`, `latest`	Meshy model to use.
`prompt`	string	Yes	-	-	-	-	Describe the object you want to generate as a 3D model.
`texturePrompt`	string	No	-	-	-	-	Provide a text description to guide the texturing process. The Style Prompt is only used when no Style Image is provided. If you upload an image, this text prompt will be ignored.
`textureImage`	file	No	-	-	-	-	Upload a reference image to guide the visual style of the generated texture. This will override the Style Prompt.
`artStyle`	string	No	`realistic`	-	-	`realistic`, `sculpture`	Describe your desired art style of the object. Note that 'Enable PBR'' should be set to false when using Art style.
`topology`	string	No	`triangle`	-	-	`triangle`, `quad`	Choose the mesh structure type used for the generated model (e.g., triangles or quads).
`enablePbr`	boolean	No	`true`	-	-	-	Generate advanced PBR maps (metallic, roughness, normal) for more realistic materials.
`poseMode`	string	No	-	-	-	``, `a-pose`, `t-pose`	Specify the pose mode for the generated model. Accepts 'a-pose', 't-pose', or empty string (default).
`shouldRemesh`	boolean	No	`false`	-	-	-	Enable remeshing to improve mesh quality and control polycount. Disable to keep the raw, unprocessed mesh.
`symmetryMode`	string	No	`auto`	-	-	`off`, `on`, `auto`	Control how symmetry is applied during model generation for balanced, even shapes.
`targetPolycount`	number	No	`30000`	100	300000	-	Specify the target number of polygons in the generated model. The actual number of polygons may deviate from the target depending on the complexity of the geometry.
`seed`	number	No	-	-	-	-	Seed for random number generator. If None, a random seed will be used.

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`image`	file	Yes	-	-	-	-	Original image used for MoGe depth estimation
`bodyMesh`	file	Yes	-	-	-	-	SAM-3D Body mesh file to align
`bodyMask`	file	No	-	-	-	-	The human mask image. If not provided, uses full image.
`objectMesh`	file	No	-	-	-	-	Optional SAM-3D Object mesh (.glb) to create combined scene
`focalLength`	number	No	-	0	-	-	Focal length from SAM-3D Body metadata. If not provided, estimated from MoGe.

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`image`	file	Yes	-	-	-	-	Input image to extract body meshes from

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`image`	file	Yes	-	-	-	-	Input image to extract objects from
`prompt`	string	No	-	-	-	-	Text prompt for auto-segmentation when no masks provided (e.g., 'chair', 'lamp')
`mask`	file_array	No	-	-	-	-	Optional list of mask URLs (one per object). If not provided, use prompt/point_prompts/box_prompts to auto-segment, or entire image will be used.
`pointPrompts`	inputs_array	No	-	-	-	-	Point prompts for auto-segmentation when no masks provided. JSON array of objects with 'x', 'y', and optional 'object_id' fields.
`boxPrompts`	inputs_array	No	-	-	-	-	Box prompts for auto-segmentation when no masks provided. JSON array of objects with 'x1', 'y1', 'x2', 'y2', and optional 'object_id' fields.
`seed`	number	No	-	0	-	-	Random seed. Leave blank to randomize the seed

Microsoft

Trellis

Model ID: model_trellis

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_trellis/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`images`	file_array	Yes	-	-	-	-	Input image(s) to generate 3D asset from
`ssGuidanceStrength`	number	No	`7.5`	0	10	-	Stage 1: Sparse Structure Generation - Guidance
`ssSamplingSteps`	number	No	`12`	1	50	-	Stage 1: Sparse Structure Generation - Steps
`slatGuidanceStrength`	number	No	`3`	0	10	-	Stage 2: Structured Latent Generation - Guidance
`slatSamplingSteps`	number	No	`12`	1	50	-	Stage 2: Structured Latent Generation - Steps
`meshSimplify`	number	No	`0.9`	0.9	0.98	-	GLB Extraction - Mesh Simplification
`textureSize`	number	No	`2048`	512	2048	-	GLB Extraction - Texture Size
`seed`	number	No	-	0	2147483647	-	Use a seed for reproducible results. Leave blank to use a random seed.

Trellis 2

Generate 3D models from your images using Trellis 2. A native 3D generative model enabling versatile and high-quality 3D asset creation.

Model ID: model_trellis-2

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_trellis-2/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`image`	file	Yes	-	-	-	-	Input image to generate 3D asset from
`resolution`	number	No	`1024`	-	-	`512`, `1024`, `1536`	Output resolution; higher is slower but more detailed
`ssGuidanceStrength`	number	No	`8.5`	0	10	-	Stage 1: Sparse Structure Generation - Guidance
`ssGuidanceRescale`	number	No	`0.7`	0	1	-	Stage 1: Sparse Structure Generation - Guidance Rescale
`ssSamplingSteps`	number	No	`20`	1	50	-	Stage 1: Sparse Structure Generation - Steps
`ssRescaleT`	number	No	`5`	1	6	-	Stage 1: Sparse Structure Generation - Rescale T
`shapeSlatGuidanceStrength`	number	No	`7.5`	0	10	-	Stage 2: Shape Structured Latent Generation - Guidance
`shapeSlatGuidanceRescale`	number	No	`0.5`	0	1	-	Stage 2: Shape Structured Latent Generation - Guidance Rescale
`shapeSlatSamplingSteps`	number	No	`20`	1	50	-	Stage 2: Shape Structured Latent Generation - Steps
`texSlatGuidanceStrength`	number	No	`2`	0	10	-	Stage 3: Texture Structured Latent Generation - Guidance
`texSlatGuidanceRescale`	number	No	`0`	0	1	-	Stage 3: Texture Structured Latent Generation - Guidance
`texSlatSamplingSteps`	number	No	`20`	1	50	-	Stage 3: Texture Structured Latent Generation - Steps
`texSlatRescaleT`	number	No	`3`	1	6	-	Stage 3: Texture Structured Latent Generation - Rescale T
`decimationTarget`	number	No	`500000`	100000	2000000	-	Target number of faces for mesh decimation
`textureSize`	number	No	`2048`	-	-	`1024`, `2048`, `4096`	Texture size for the generated 3D model
`remesh`	boolean	No	`true`	-	-	-	Enable remeshing
`remeshBand`	number	No	`1`	1	4	-	Remesh band parameter
`seed`	number	No	-	0	2147483647	-	Use a seed for reproducible results. Leave blank to use a random seed.

Trellis 2 Retexture

Retexture 3D models with Trellis 2. Apply new textures to existing meshes using a reference image for style guidance.

Model ID: model_trellis-2-retexture

Capabilities: 3d23d

LLM Markdown: https://app.scenario.com/api/models/model_trellis-2-retexture/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`model`	file	Yes	-	-	-	-	Input mesh
`image`	file	Yes	-	-	-	-	Reference image for retexturing
`resolution`	number	No	`1024`	-	-	`512`, `1024`	Output resolution; higher is slower but more detailed
`textureSize`	number	No	`2048`	-	-	`1024`, `2048`, `4096`	Texture size for the retextured 3D model
`texSlatGuidanceStrength`	number	No	`1`	0	10	-	Guidance strength
`texSlatGuidanceRescale`	number	No	`0`	0	1	-	Texture generation guidance rescale
`texSlatSamplingSteps`	number	No	`12`	1	50	-	Texture generation sampling steps
`texSlatRescaleT`	number	No	`3`	1	6	-	Texture generation rescale T
`seed`	number	No	-	0	2147483647	-	Use a seed for reproducible results. Leave blank to use a random seed.

Tencent

Hunyuan 3D - Fast

Model ID: model_hunyuan-3d-fast

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_hunyuan-3d-fast/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`image`	file	Yes	-	-	-	-	Front view to transform in 3D, should remove background to get better results
`paint`	boolean	No	`true`	-	-	-	Paint the mesh with texture.
`steps`	number	No	`30`	10	50	-	The number of steps for the generation
`guidanceScale`	number	No	`5`	1	10	-	Higher values adhere more closely to the input image, while lower values allow for more creative freedom.
`targetFaceNum`	number	No	`40000`	100	1000000	-	The number of faces to target for the generation
`seed`	number	No	-	0	2147483647	-	Use a seed for reproducible results. Leave blank to use a random seed.

Hunyuan 3D 2.0

Model ID: model_hunyuan-3d

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_hunyuan-3d/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`image`	file	Yes	-	-	-	-	Front view to transform in 3D, should remove background to get better results
`paint`	boolean	No	`true`	-	-	-	Paint the mesh with texture.
`steps`	number	No	`30`	10	50	-	The number of steps for the generation
`guidanceScale`	number	No	`5`	1	10	-	Higher values adhere more closely to the input image, while lower values allow for more creative freedom.
`targetFaceNum`	number	No	`40000`	100	1000000	-	The number of faces to target for the generation
`seed`	number	No	-	0	2147483647	-	Use a seed for reproducible results. Leave blank to use a random seed.

Hunyuan 3D 2.0 - Multi View

Model ID: model_hunyuan-3d-mv

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_hunyuan-3d-mv/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`image`	file	Yes	-	-	-	-	Front view to transform in 3D, should remove background to get better results
`backImage`	file	No	-	-	-	-	Back view to transform in 3D, should remove background to get better results
`leftImage`	file	No	-	-	-	-	Left view to transform in 3D, should remove background to get better results
`rightImage`	file	No	-	-	-	-	Right view to transform in 3D, should remove background to get better results
`paint`	boolean	No	`true`	-	-	-	Paint the mesh with texture.
`steps`	number	No	`30`	10	50	-	The number of steps for the generation
`guidanceScale`	number	No	`5`	1	10	-	Higher values adhere more closely to the input image, while lower values allow for more creative freedom.
`targetFaceNum`	number	No	`40000`	100	1000000	-	The number of faces to target for the generation
`seed`	number	No	-	0	2147483647	-	Use a seed for reproducible results. Leave blank to use a random seed.

Hunyuan 3D 2.1

Model ID: model_hunyuan-3d-v2-1

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_hunyuan-3d-v2-1/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`image`	file	Yes	-	-	-	-	Front view to transform in 3D, should remove background to get better results
`paint`	boolean	No	`true`	-	-	-	Paint the mesh with texture.
`steps`	number	No	`30`	10	100	-	The number of steps for the generation
`guidanceScale`	number	No	`5`	1	10	-	Higher values adhere more closely to the input image, while lower values allow for more creative freedom.
`targetFaceNum`	number	No	`40000`	100	1000000	-	The number of faces to target for the generation
`seed`	number	No	-	0	2147483647	-	Use a seed for reproducible results. Leave blank to use a random seed.

Hunyuan 3D 3.0 Pro

Model ID: model_hunyuan-3d-pro-i23d

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_hunyuan-3d-pro-i23d/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`image`	file	Yes	-	-	-	-	Input image. Required if prompt not provided. Size: 128-5000 pixels per side, max 8MB. Formats: jpg, png, jpeg, webp.
`faceCount`	number	No	`500000`	40000	1500000	-	Target number of faces in the generated 3D mesh. Higher values produce finer detail. This parameter is ignored when 'Optimized Mesh' mode is enabled.
`generateType`	string	No	`Normal`	-	-	`Normal`, `LowPoly`, `Geometry`	Standard: textured mesh. Optimized Mesh: simplified mesh. Mesh Only: no textures.
`enablePbr`	boolean	No	`false`	-	-	-	Enables Physically Based Rendering (PBR) materials for realistic lighting.
`polygonType`	string	No	`triangle`	-	-	`triangle`, `quadrilateral`	Polygon grid format. Tris for real-time engines; Quads for sculpting.

Hunyuan 3D 3.0 Pro (Multiview)

Model ID: model_hunyuan-3d-pro-multiview

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_hunyuan-3d-pro-multiview/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`frontImage`	file	Yes	-	-	-	-	Front view image for multiview generation. Size: 128-5000 pixels per side, max 8MB.
`leftImage`	file	No	-	-	-	-	Left view image for multiview generation. Size: 128-5000 pixels per side, max 8MB
`rightImage`	file	No	-	-	-	-	Right view image for multiview generation. Size: 128-5000 pixels per side, max 8MB.
`backImage`	file	No	-	-	-	-	Back view image for multiview generation. Size: 128-5000 pixels per side, max 8MB.
`faceCount`	number	No	`500000`	40000	1500000	-	Target number of faces in the generated 3D mesh. Higher values produce finer detail. This parameter is ignored when 'Optimized Mesh' mode is enabled.
`generateType`	string	No	`Normal`	-	-	`Normal`, `LowPoly`, `Geometry`	Standard: textured mesh. Optimized Mesh: simplified mesh. Mesh Only: no textures.
`enablePbr`	boolean	No	`false`	-	-	-	Enables Physically Based Rendering (PBR) materials for realistic lighting.
`polygonType`	string	No	`triangle`	-	-	`triangle`, `quadrilateral`	Polygon grid format. Tris for real-time engines; Quads for sculpting.

Hunyuan 3D 3.0 Pro (Sketch)

Model ID: model_hunyuan-3d-pro-sketch

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_hunyuan-3d-pro-sketch/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`prompt`	string	Yes	-	-	-	-	Text prompt for 3D content generation.
`image`	file	Yes	-	-	-	-	Input sketch image. Size: 128-5000 pixels per side, max 8MB.
`faceCount`	number	No	`500000`	40000	1500000	-	Target number of faces in the generated 3D mesh. Higher values produce finer detail.
`enablePbr`	boolean	No	`false`	-	-	-	Enables Physically Based Rendering (PBR) materials for realistic lighting.
`polygonType`	string	No	`triangle`	-	-	`triangle`, `quadrilateral`	Polygon grid format. Tris for real-time engines; Quads for sculpting.

Hunyuan 3D 3.1 Pro

Hunyuan 3D 3.1 (Pro) by Tencent is a state-of-the-art 10B parameter image-to-3D model with 1536³ resolution, hierarchical DiT carving, and enhanced texture color accuracy.

Model ID: model_hunyuan-3d-pro-3-1-i23d

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_hunyuan-3d-pro-3-1-i23d/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`image`	file	Yes	-	-	-	-	Input image. Required if prompt not provided. Size: 128-5000 pixels per side, max 8MB. Formats: jpg, png, jpeg, webp.
`faceCount`	number	No	`500000`	40000	1500000	-	Target number of faces in the generated 3D mesh. Higher values produce finer detail.
`generateType`	string	No	`Normal`	-	-	`Normal`, `Geometry`	Standard: textured mesh. Mesh Only: no textures.
`enablePbr`	boolean	No	`false`	-	-	-	Enables Physically Based Rendering (PBR) materials for realistic lighting.

Hunyuan 3D 3.1 Pro (Multiview)

Hunyuan 3D 3.1 (Multiview) reconciles up to 8 reference images to produce symmetric 3D assets from 120 credits.

Model ID: model_hunyuan-3d-pro-3-1-multiview

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_hunyuan-3d-pro-3-1-multiview/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`frontImage`	file	Yes	-	-	-	-	Front view image for multiview generation. Size: 128-5000 pixels per side, max 8MB.
`leftImage`	file	No	-	-	-	-	Left view image for multiview generation. Size: 128-5000 pixels per side, max 8MB
`rightImage`	file	No	-	-	-	-	Right view image for multiview generation. Size: 128-5000 pixels per side, max 8MB.
`backImage`	file	No	-	-	-	-	Back view image for multiview generation. Size: 128-5000 pixels per side, max 8MB.
`topImage`	file	No	-	-	-	-	Top view image for multiview generation. Size: 128-5000 pixels per side, max 8MB.
`bottomImage`	file	No	-	-	-	-	Bottom view image for multiview generation. Size: 128-5000 pixels per side, max 8MB.
`leftFrontImage`	file	No	-	-	-	-	Left front 45° view image for multiview generation. Size: 128-5000 pixels per side, max 8MB.
`rightFrontImage`	file	No	-	-	-	-	Right front 45° view image for multiview generation. Size: 128-5000 pixels per side, max 8MB.
`faceCount`	number	No	`500000`	40000	1500000	-	Target number of faces in the generated 3D mesh. Higher values produce finer detail.
`generateType`	string	No	`Normal`	-	-	`Normal`, `Geometry`	Standard: textured mesh. Mesh Only: no textures.
`enablePbr`	boolean	No	`false`	-	-	-	Enables Physically Based Rendering (PBR) materials for realistic lighting.

Hunyuan 3D 3.1 Pro (Sketch)

Hunyuan 3D 3.1 (Sketch) transforms hand-drawn line art and sketches into textured 3D meshes from 105 credits.

Model ID: model_hunyuan-3d-pro-3-1-sketch

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_hunyuan-3d-pro-3-1-sketch/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`prompt`	string	Yes	-	-	-	-	Text prompt for 3D content generation.
`image`	file	Yes	-	-	-	-	Input sketch image. Size: 128-5000 pixels per side, max 8MB.
`faceCount`	number	No	`500000`	40000	1500000	-	Target number of faces in the generated 3D mesh. Higher values produce finer detail.
`enablePbr`	boolean	No	`false`	-	-	-	Enables Physically Based Rendering (PBR) materials for realistic lighting.

Hunyuan 3D Multi View - Fast

Model ID: model_hunyuan-3d-mv-fast

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_hunyuan-3d-mv-fast/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`image`	file	Yes	-	-	-	-	Front view to transform in 3D, should remove background to get better results
`backImage`	file	No	-	-	-	-	Back view to transform in 3D, should remove background to get better results
`leftImage`	file	No	-	-	-	-	Left view to transform in 3D, should remove background to get better results
`rightImage`	file	No	-	-	-	-	Right view to transform in 3D, should remove background to get better results
`paint`	boolean	No	`true`	-	-	-	Paint the mesh with texture.
`steps`	number	No	`30`	10	50	-	The number of steps for the generation
`guidanceScale`	number	No	`5`	1	10	-	Higher values adhere more closely to the input image, while lower values allow for more creative freedom.
`targetFaceNum`	number	No	`40000`	100	1000000	-	The number of faces to target for the generation
`seed`	number	No	-	0	2147483647	-	Use a seed for reproducible results. Leave blank to use a random seed.

Hunyuan 3D Part

Model ID: model_hunyuan-3d-part

Capabilities: 3d23d

LLM Markdown: https://app.scenario.com/api/models/model_hunyuan-3d-part/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`mesh`	file	Yes	-	-	-	-	Upload a GLB file of the 3D model you want to generate a part of.
`postprocess`	boolean	No	`true`	-	-	-	Post-processingwill merge the small parts according to the threshold. The smaller the threshold, the more parts will be merged.
`postprocessingThreshold`	number	No	`0.95`	0	1	-	Threshold for postprocessing the segmentation mask
`seed`	number	No	-	-	-	-	Use a seed for reproducible results. Leave blank to use a random seed.

Hunyuan Polygen 1.5

Hunyuan Polygen 1.5 by Tencent is an art-grade AI retopology model that converts high-poly 3D meshes into clean, production-ready low-poly assets with quad or triangle topology.

Model ID: model_tencent-smarttopology

Capabilities: 3d23d

LLM Markdown: https://app.scenario.com/api/models/model_tencent-smarttopology/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`file3d`	file	Yes	-	-	-	-	Source 3D file model link. Supported formats: GLB, OBJ. Maximum file size: 200MB.
`polygonType`	string	No	`triangle`	-	-	`triangle`, `quadrilateral`	Polygon type, indicates model surface composition. Options: triangle, quadrilateral.
`faceLevel`	string	No	`medium`	-	-	`high`, `medium`, `low`	Reduction level for polygon count. Options: high (minimal reduction), medium (balanced), low (aggressive reduction).

Tencent Texture Edit

AI-powered texture editing for 3D models. Apply textures from text prompts or reference images to FBX models. Supports PBR (Physically Based Rendering) when using prompts.

Model ID: model_tencent-texture-edit

Capabilities: 3d23d

LLM Markdown: https://app.scenario.com/api/models/model_tencent-texture-edit/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`file3d`	file	Yes	-	-	-	-	Source 3D model file. Supported format: FBX. Maximum: 100,000 faces.
`image`	file	No	-	-	-	-	Reference image for texture style. Resolution: 128-4096px. Cannot be used with texture prompt.
`prompt`	string	No	-	-	-	-	Texture description prompt. Cannot be used with reference image.

Tencent UV Unwrapping

AI-powered UV unwrapping for 3D models. Generates clean UV maps for FBX, OBJ, and GLB models with up to 30,000 faces.

Model ID: model_tencent-uv-unwrapping

Capabilities: 3d23d

LLM Markdown: https://app.scenario.com/api/models/model_tencent-uv-unwrapping/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`file3d`	file	Yes	-	-	-	-	Source 3D model file. Maximum: 30,000 faces.

Tripo AI

Tripo 2.5

Tripo V2.5 Image to 3D model.

Model ID: model_tripo-v2-5-image-to-3d

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_tripo-v2-5-image-to-3d/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`image`	file	Yes	-	-	-	-	Single image for image-to-model.
`texture`	boolean	No	`true`	-	-	-	Enable texturing. Set to false for a model without textures.
`textureQuality`	string	No	`standard`	-	-	`standard`, `detailed`	Texture quality level. 'Detailed' gives HD quality textures.
`textureAlignment`	string	No	`original_image`	-	-	`original_image`, `geometry`	Determines the prioritization of texture alignment in the 3D model.
`textureSeed`	number	No	-	-	-	-	Random seed for texture generation. Using the same seed will produce identical textures.
`orientation`	string	No	`default`	-	-	`default`, `align_image`	Set orientation to 'Align Image' will automatically rotate the model to align the original image.
`pbr`	boolean	No	`true`	-	-	-	Enable PBR generation. Default value is True. If this option is set to True, texture parameters will be ignored.
`faceLimit`	number	No	-	-	-	-	Limits the number of faces on the output model. If this option is not set, the face limit will be adaptively determined.
`autoSize`	boolean	No	`false`	-	-	-	Automatically scale the model to real-world dimensions, with the unit in meters.
`quad`	boolean	No	`false`	-	-	-	Set True to enable quad mesh output. If Quad is True and Face Limit is not set, the default Face Limit will be 10000.
`smartLowPoly`	boolean	No	`false`	-	-	-	Generate hand-crafted low-poly topology.
`generateParts`	boolean	No	`false`	-	-	-	Generate segmented 3D model parts. Incompatible with texture, pbr, and quad.
`seed`	number	No	-	-	-	-	Random seed for model generation. The seed controls the geometry generation process, ensuring identical models when the same seed is used.

Tripo 2.5 Multi View

Tripo V2.5 Multiview (front + left/back/right) to 3D model.

Model ID: model_tripo-v2-5-multiview-to-3d

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_tripo-v2-5-multiview-to-3d/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`frontImage`	file	Yes	-	-	-	-	Front view image (0 deg). Required for multiview mode. Provide with at least one of left, back, right.
`leftImage`	file	No	-	-	-	-	Left view image (90 deg — character's left arm side).
`backImage`	file	No	-	-	-	-	Back view image (180 deg).
`rightImage`	file	No	-	-	-	-	Right view image (270 deg).
`texture`	boolean	No	`true`	-	-	-	Enable texturing. Set to false for a model without textures.
`textureQuality`	string	No	`standard`	-	-	`standard`, `detailed`	Texture quality level. 'Detailed' gives HD quality textures.
`textureAlignment`	string	No	`original_image`	-	-	`original_image`, `geometry`	Determines the prioritization of texture alignment in the 3D model.
`textureSeed`	number	No	-	-	-	-	Random seed for texture generation. Using the same seed will produce identical textures.
`orientation`	string	No	`default`	-	-	`default`, `align_image`	Set orientation to 'Align Image' will automatically rotate the model to align the original image.
`pbr`	boolean	No	`true`	-	-	-	Enable PBR generation. Default value is True. If this option is set to True, texture parameters will be ignored.
`faceLimit`	number	No	-	-	-	-	Limits the number of faces on the output model. If this option is not set, the face limit will be adaptively determined.
`autoSize`	boolean	No	`false`	-	-	-	Automatically scale the model to real-world dimensions, with the unit in meters.
`quad`	boolean	No	`false`	-	-	-	Set True to enable quad mesh output. If Quad is True and Face Limit is not set, the default Face Limit will be 10000.
`smartLowPoly`	boolean	No	`false`	-	-	-	Generate hand-crafted low-poly topology.
`generateParts`	boolean	No	`false`	-	-	-	Generate segmented 3D model parts. Incompatible with texture, pbr, and quad.
`seed`	number	No	-	-	-	-	Random seed for model generation. The seed controls the geometry generation process, ensuring identical models when the same seed is used.

Tripo 3.0

Tripo V3.0 Image to 3D model.

Model ID: model_tripo-v3-0-image-to-3d

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_tripo-v3-0-image-to-3d/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`image`	file	Yes	-	-	-	-	Single image for image-to-model.
`texture`	boolean	No	`true`	-	-	-	Enable texturing. Set to false for a model without textures.
`textureQuality`	string	No	`standard`	-	-	`standard`, `detailed`	Texture quality level. 'Detailed' gives HD quality textures.
`textureAlignment`	string	No	`original_image`	-	-	`original_image`, `geometry`	Determines the prioritization of texture alignment in the 3D model.
`geometryQuality`	string	No	`standard`	-	-	`standard`, `detailed`	Geometry quality level. 'Detailed' gives HD quality geometry.
`textureSeed`	number	No	-	-	-	-	Random seed for texture generation. Using the same seed will produce identical textures.
`orientation`	string	No	`default`	-	-	`default`, `align_image`	Set orientation to 'Align Image' will automatically rotate the model to align the original image.
`pbr`	boolean	No	`true`	-	-	-	Enable PBR generation. Default value is True. If this option is set to True, texture parameters will be ignored.
`faceLimit`	number	No	-	-	-	-	Limits the number of faces on the output model. If this option is not set, the face limit will be adaptively determined.
`autoSize`	boolean	No	`false`	-	-	-	Automatically scale the model to real-world dimensions, with the unit in meters.
`quad`	boolean	No	`false`	-	-	-	Set True to enable quad mesh output.
`smartLowPoly`	boolean	No	`false`	-	-	-	Generate hand-crafted low-poly topology.
`generateParts`	boolean	No	`false`	-	-	-	Generate segmented 3D model parts. Incompatible with texture, pbr, and quad.
`seed`	number	No	-	-	-	-	Random seed for model generation. The seed controls the geometry generation process, ensuring identical models when the same seed is used.

Tripo 3.0 Multi View

Tripo V3.0 Multiview (front + left/back/right) to 3D model.

Model ID: model_tripo-v3-0-multiview-to-3d

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_tripo-v3-0-multiview-to-3d/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`frontImage`	file	Yes	-	-	-	-	Front view image (0 deg). Required for multiview mode. Provide with at least one of left, back, right.
`leftImage`	file	No	-	-	-	-	Left view image (90 deg — character's left arm side).
`backImage`	file	No	-	-	-	-	Back view image (180 deg).
`rightImage`	file	No	-	-	-	-	Right view image (270 deg).
`texture`	boolean	No	`true`	-	-	-	Enable texturing. Set to false for a model without textures.
`textureQuality`	string	No	`standard`	-	-	`standard`, `detailed`	Texture quality level. 'Detailed' gives HD quality textures.
`textureAlignment`	string	No	`original_image`	-	-	`original_image`, `geometry`	Determines the prioritization of texture alignment in the 3D model.
`geometryQuality`	string	No	`standard`	-	-	`standard`, `detailed`	Geometry quality level. 'Detailed' gives HD quality geometry.
`textureSeed`	number	No	-	-	-	-	Random seed for texture generation. Using the same seed will produce identical textures.
`orientation`	string	No	`default`	-	-	`default`, `align_image`	Set orientation to 'Align Image' will automatically rotate the model to align the original image.
`pbr`	boolean	No	`true`	-	-	-	Enable PBR generation. Default value is True. If this option is set to True, texture parameters will be ignored.
`faceLimit`	number	No	-	-	-	-	Limits the number of faces on the output model. If this option is not set, the face limit will be adaptively determined.
`autoSize`	boolean	No	`false`	-	-	-	Automatically scale the model to real-world dimensions, with the unit in meters.
`quad`	boolean	No	`false`	-	-	-	Set True to enable quad mesh output. If Quad is True and Face Limit is not set, the default Face Limit will be 10000.
`smartLowPoly`	boolean	No	`false`	-	-	-	Generate hand-crafted low-poly topology.
`generateParts`	boolean	No	`false`	-	-	-	Generate segmented 3D model parts. Incompatible with texture, pbr, and quad.
`seed`	number	No	-	-	-	-	Random seed for model generation. The seed controls the geometry generation process, ensuring identical models when the same seed is used.

Tripo 3.0 Texturing

Tripo V3.0 Texture service - texture existing 3D models with text prompts, image prompts, or multiview images. Supports standard/HD quality and style reference.

Model ID: model_tripo-v3-0-texturing

Capabilities: 3d23d

LLM Markdown: https://app.scenario.com/api/models/model_tripo-v3-0-texturing/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`model`	file	Yes	-	-	-	-	Source 3D model.
`texture`	boolean	No	`true`	-	-	-	Enable texture generation.
`pbr`	boolean	No	`false`	-	-	-	PBR generates material maps (albedo, metallic, roughness, normal). When enabled, texture options are ignored.
`textureQuality`	string	No	`standard`	-	-	`standard`, `detailed`	Texture quality level. 'Detailed' gives HD quality textures.
`textureAlignment`	string	No	`original_image`	-	-	`original_image`, `geometry`	Texture alignment strategy.
`textureSeed`	number	No	-	-	-	-	Random seed for texture generation.
`prompt`	string	No	-	-	-	-	Text description for texture generation. Mutually exclusive with image prompt and multiview images.
`imagePrompt`	file	No	-	-	-	-	Single image used as texture prompt. Mutually exclusive with text prompt.
`styleImage`	file	No	-	-	-	-	Reference image for artistic style. Requires prompt or image prompt to also be set.
`bake`	boolean	No	`true`	-	-	-	Bake textures onto the model.

Tripo 3.1

Tripo V3.1 Image to 3D model.

Model ID: model_tripo-v3-1-image-to-3d

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_tripo-v3-1-image-to-3d/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`image`	file	Yes	-	-	-	-	Single image for image-to-model.
`texture`	boolean	No	`true`	-	-	-	Enable texturing. Set to false for a model without textures.
`textureQuality`	string	No	`standard`	-	-	`standard`, `detailed`	Texture quality level. 'Detailed' gives HD quality textures.
`textureAlignment`	string	No	`original_image`	-	-	`original_image`, `geometry`	Determines the prioritization of texture alignment in the 3D model.
`geometryQuality`	string	No	`standard`	-	-	`standard`, `detailed`	Geometry quality level. 'Detailed' gives HD quality geometry.
`textureSeed`	number	No	-	-	-	-	Random seed for texture generation. Using the same seed will produce identical textures.
`orientation`	string	No	`default`	-	-	`default`, `align_image`	Set orientation to 'Align Image' will automatically rotate the model to align the original image.
`pbr`	boolean	No	`true`	-	-	-	Enable PBR generation. Default value is True. If this option is set to True, texture parameters will be ignored.
`faceLimit`	number	No	-	-	-	-	Limits the number of faces on the output model. If this option is not set, the face limit will be adaptively determined.
`autoSize`	boolean	No	`false`	-	-	-	Automatically scale the model to real-world dimensions, with the unit in meters.
`quad`	boolean	No	`false`	-	-	-	Set True to enable quad mesh output.
`smartLowPoly`	boolean	No	`false`	-	-	-	Generate hand-crafted low-poly topology.
`generateParts`	boolean	No	`false`	-	-	-	Generate segmented 3D model parts. Incompatible with texture, pbr, and quad.
`seed`	number	No	-	-	-	-	Random seed for model generation. The seed controls the geometry generation process, ensuring identical models when the same seed is used.

Tripo 3.1 Multi View

Tripo V3.1 Multiview (front + left/back/right) to 3D model.

Model ID: model_tripo-v3-1-multiview-to-3d

Capabilities: img23d

LLM Markdown: https://app.scenario.com/api/models/model_tripo-v3-1-multiview-to-3d/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`frontImage`	file	Yes	-	-	-	-	Front view image (0 deg). Required for multiview mode. Provide with at least one of left, back, right.
`leftImage`	file	No	-	-	-	-	Left view image (90 deg — character's left arm side).
`backImage`	file	No	-	-	-	-	Back view image (180 deg).
`rightImage`	file	No	-	-	-	-	Right view image (270 deg).
`texture`	boolean	No	`true`	-	-	-	Enable texturing. Set to false for a model without textures.
`textureQuality`	string	No	`standard`	-	-	`standard`, `detailed`	Texture quality level. 'Detailed' gives HD quality textures.
`textureAlignment`	string	No	`original_image`	-	-	`original_image`, `geometry`	Determines the prioritization of texture alignment in the 3D model.
`geometryQuality`	string	No	`standard`	-	-	`standard`, `detailed`	Geometry quality level. 'Detailed' gives HD quality geometry.
`textureSeed`	number	No	-	-	-	-	Random seed for texture generation. Using the same seed will produce identical textures.
`orientation`	string	No	`default`	-	-	`default`, `align_image`	Set orientation to 'Align Image' will automatically rotate the model to align the original image.
`pbr`	boolean	No	`true`	-	-	-	Enable PBR generation. Default value is True. If this option is set to True, texture parameters will be ignored.
`faceLimit`	number	No	-	-	-	-	Limits the number of faces on the output model. If this option is not set, the face limit will be adaptively determined.
`autoSize`	boolean	No	`false`	-	-	-	Automatically scale the model to real-world dimensions, with the unit in meters.
`quad`	boolean	No	`false`	-	-	-	Set True to enable quad mesh output. If Quad is True and Face Limit is not set, the default Face Limit will be 10000.
`smartLowPoly`	boolean	No	`false`	-	-	-	Generate hand-crafted low-poly topology.
`generateParts`	boolean	No	`false`	-	-	-	Generate segmented 3D model parts. Incompatible with texture, pbr, and quad.
`seed`	number	No	-	-	-	-	Random seed for model generation. The seed controls the geometry generation process, ensuring identical models when the same seed is used.

Tripo Retopology

Tripo SmartLowpoly - Convert high-poly 3D meshes to clean low-poly with optional quad output.

Model ID: model_tripo-retopology

Capabilities: 3d23d

LLM Markdown: https://app.scenario.com/api/models/model_tripo-retopology/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`model`	file	Yes	-	-	-	-	Source 3D model.
`faceLimit`	number	No	-	1000	20000	-	Target face count. Adaptive if not set.
`quad`	boolean	No	`false`	-	-	-	Output quad faces instead of triangles.
`bake`	boolean	No	`true`	-	-	-	Bake textures onto lowpoly mesh.

Tripo Rigging v1

Tripo Rigging v1.0 - Rig 3D models with optional animation retargeting. 100+ biped animations.

Model ID: model_tripo-rigging-v1

Capabilities: 3d23d

LLM Markdown: https://app.scenario.com/api/models/model_tripo-rigging-v1/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`model`	file	Yes	-	-	-	-	Source 3D file. Supported formats: GLB, OBJ, FBX, STL.
`rigType`	string	No	`biped`	-	-	`biped`, `quadruped`, `hexapod`, `octopod`, `avian`, `serpentine`, `aquatic`, `others`	Skeleton type for rigging.
`animation`	string	No	-	-	-	``, `biped:idle`, `biped:walk`, `biped:run`, `biped:dive`, `biped:climb`, `biped:jump`, `biped:slash`, `biped:shoot`, `biped:hurt`, `biped:fall`, `biped:turn`, `quadruped:walk`, `hexapod:walk`, `octopod:walk`, `serpentine:march`, `aquatic:march`	Animation to retarget (e.g. biped:walk, biped:dance_01, quadruped:walk). Leave empty for rigging only. 100+ biped animations available in v1.
`includeRiggedModel`	boolean	No	`false`	-	-	-	Include the rigged model (without animations) in output. When animation is set, only retarget file is returned by default.

Tripo Rigging v2.5

Tripo Rigging v2.5 - Rig 3D models with optional animation retargeting. Recommended for non-biped rigs.

Model ID: model_tripo-rigging-v2-5

Capabilities: 3d23d

LLM Markdown: https://app.scenario.com/api/models/model_tripo-rigging-v2-5/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`model`	file	Yes	-	-	-	-	Source 3D file. Supported formats: GLB, OBJ, FBX, STL.
`rigType`	string	No	`biped`	-	-	`biped`, `quadruped`, `hexapod`, `octopod`, `avian`, `serpentine`, `aquatic`, `others`	Skeleton type for rigging.
`animation`	string	No	-	-	-	``, `biped:idle`, `biped:walk`, `biped:run`, `biped:dive`, `biped:climb`, `biped:jump`, `biped:slash`, `biped:shoot`, `biped:hurt`, `biped:fall`, `biped:turn`, `quadruped:walk`, `hexapod:walk`, `octopod:walk`, `serpentine:march`, `aquatic:march`	Animation to retarget (e.g. biped:walk, biped:run, quadruped:walk). Leave empty for rigging only.
`includeRiggedModel`	boolean	No	`false`	-	-	-	Include the rigged model (without animations) in output. When animation is set, only retarget file is returned by default.

Tripo Stylization

Tripo Stylization - Apply LEGO, voxel or Voronoi style to 3D models.

Model ID: model_tripo-stylization

Capabilities: 3d23d

LLM Markdown: https://app.scenario.com/api/models/model_tripo-stylization/markdown

Parameter	Type	Required	Default	Min	Max	Allowed Values	Description
`model`	file	Yes	-	-	-	-	Source 3D model.
`style`	string	Yes	`lego`	-	-	`lego`, `voxel`, `voronoi`	Stylization type to apply.

Models

Scenario

Academia

Deemos Technology

Hitem3D

Meshy

Meta

Microsoft

Tencent

Tripo AI

Scenario

Scenario Voxels

Academia

PartCrafter

Ultrashape 1.0

Deemos Technology

Rodin Hyper3D

Rodin Hyper3D Gen-2

Rodin Hyper3D HighPack

Hitem3D

Sparc3D

Sparc3D 2.0

Sparc3D 2.0 Portrait

Sparc3D Portrait

Meshy

Meshy Image-to-3D

Meshy Remesh

Meshy Retexture

Meshy Rigging

Meshy Text-to-3D

Meta

SAM3D Align

SAM3D Body

SAM3D Objects

Microsoft

Trellis

Trellis 2

Trellis 2 Retexture

Tencent

Hunyuan 3D - Fast

Hunyuan 3D 2.0

Hunyuan 3D 2.0 - Multi View

Hunyuan 3D 2.1

Hunyuan 3D 3.0 Pro

Hunyuan 3D 3.0 Pro (Multiview)

Hunyuan 3D 3.0 Pro (Sketch)

Hunyuan 3D 3.1 Pro

Hunyuan 3D 3.1 Pro (Multiview)

Hunyuan 3D 3.1 Pro (Sketch)

Hunyuan 3D Multi View - Fast

Hunyuan 3D Part

Hunyuan Polygen 1.5

Tencent Texture Edit

Tencent UV Unwrapping

Tripo AI

Tripo 2.5

Tripo 2.5 Multi View

Tripo 3.0

Tripo 3.0 Multi View

Tripo 3.0 Texturing

Tripo 3.1

Tripo 3.1 Multi View

Tripo Retopology

Tripo Rigging v1

Tripo Rigging v2.5

Tripo Stylization