You're looking at a specific version of this model. Jump to the model overview.

fishwowater /trellis2:52e1ad68

Input schema

The fields you can use to run this model with an API. If you don’t give a value for a field its default value will be used.

Field Type Default value Description
image
string
Input image to generate 3D asset from
seed
integer
42
Random seed for generation
randomize_seed
boolean
False
Randomize seed
preprocess_image
boolean
True
Preprocess image (remove background and crop)
return_no_background
boolean
False
Return the preprocessed image without background
generate_video
boolean
True
Generate rendered video preview
generate_model
boolean
True
Generate 3D model file (GLB)
pipeline_type
None
1024_cascade
Pipeline type for quality/speed tradeoff
sparse_structure_steps
integer
12

Min: 1

Max: 50

Sparse structure sampling steps (more steps = better quality, slower)
sparse_structure_guidance_strength
number
7.5

Max: 15

Sparse structure guidance strength
sparse_structure_guidance_rescale
number
0.7

Max: 1

Sparse structure guidance rescale
sparse_structure_rescale_t
number
5

Min: 1

Max: 6

Sparse structure rescale T
shape_slat_steps
integer
12

Min: 1

Max: 50

Shape SLat sampling steps
shape_slat_guidance_strength
number
7.5

Max: 15

Shape SLat guidance strength
shape_slat_guidance_rescale
number
0.5

Max: 1

Shape SLat guidance rescale
shape_slat_rescale_t
number
3

Min: 1

Max: 6

Shape SLat rescale T
tex_slat_steps
integer
12

Min: 1

Max: 50

Texture SLat sampling steps
tex_slat_guidance_strength
number
1

Max: 15

Texture SLat guidance strength
tex_slat_guidance_rescale
number
0

Max: 1

Texture SLat guidance rescale
tex_slat_rescale_t
number
3

Min: 1

Max: 6

Texture SLat rescale T
decimation_target
integer
1000000

Min: 100000

Max: 2000000

Target number of faces for decimation (only used if generate_model=True)
texture_size
integer
4096

Min: 1024

Max: 8192

Texture size for GLB export (only used if generate_model=True)

Output schema

The shape of the response you’ll get when you run this model with an API.

Schema
{'description': 'Output model for predictions',
 'properties': {'model_file': {'format': 'uri',
                               'nullable': True,
                               'title': 'Model File',
                               'type': 'string'},
                'no_background_image': {'format': 'uri',
                                        'nullable': True,
                                        'title': 'No Background Image',
                                        'type': 'string'},
                'video': {'format': 'uri',
                          'nullable': True,
                          'title': 'Video',
                          'type': 'string'}},
 'title': 'PredictOutput',
 'type': 'object'}