You're looking at a specific version of this model. Jump to the model overview.
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 |
|---|---|---|---|
| job_id |
string
|
Job ID for the model. Randomly generated if not provided
|
|
| image |
string
|
Base image for the model
|
|
| subject |
string
|
Subject of the image
|
|
| placed_on |
string
|
Where the subject is placed
|
|
| background |
string
|
Background description
|
|
| camera_orientation |
string
|
Camera orientation
|
|
| camera_angle |
string
|
Camera angle
|
|
| main_light_direction |
string
|
Main light direction
|
|
| proposed_background |
string
|
Proposed background
|
|
| proposed_brightness |
number
|
Proposed brightness
|
|
| prompt_suffix |
string
|
,high quality, medium format, f/32, realistic, cinematic, RAW
|
Prompt suffix
|
| prompt_prefix |
string
|
photograph
|
Starting prompt for the model. Gets added with other variables as needed
|
| negative_prompt |
string
|
|
Negative prompt for the model. Not necessary, but can be used to guide the model
|
| override_prompt |
string
|
|
Override prompt for the model. Not necessary, but can be used to guide the model
|
| background_brightness |
number
|
0.5
Min: -1 Max: 1 |
Background brightness for the model
|
| force_center |
boolean
|
False
|
Force the subject to be centered
|
| seed |
integer
|
-1
|
Random seed for the model. Leave as -1 for random seed
|
| height |
integer
|
768
|
Height of the output image. Must be divisible by 32, but this is handled automatically
|
| width |
integer
|
1024
|
Width of the output image. Must be divisible by 32, but this is handled automatically
|
| final_step_upscale_4x |
boolean
|
False
|
Final step upscale 4x for the model
|
| debug_mode |
boolean
|
False
|
Debug mode for the model
|
| object_padding |
number
|
0.1
Max: 1 |
Object padding for the model. Percent of the object width
|
| first_pass_steps |
integer
|
10
Min: 10 Max: 50 |
Number of steps for the first pass
|
| first_pass_i2i_strength |
number
|
1
Max: 1 |
I2I strength for the first pass. Must be 1.0 unless you feel like experimenting
|
| first_pass_pag_scale |
number
|
2
Min: 0.5 Max: 5 |
PAG scale for the first pass. 1 or 2 seem to work best
|
| first_pass_guidance_scale |
number
|
4
Min: 1 Max: 14 |
Guidance scale for the first pass
|
| first_pass_outpaint_strength |
number
|
1
Max: 1 |
Outpaint strength for the first pass
|
| first_pass_depth_strength |
number
|
0
Max: 1 |
Depth strength for the first pass. For the first pass we want more flexibility, so 0.0 is a good starting point
|
| first_pass_line_strength |
number
|
1
Max: 1 |
Line strength for the first pass
|
| skip_second_pass |
boolean
|
False
|
Skip the second pass
|
| second_pass_steps |
integer
|
10
Min: 10 Max: 50 |
Number of steps for the second pass
|
| second_pass_i2i_strength |
number
|
0.5
Max: 1 |
I2I strength for the second pass
|
| second_pass_pag_scale |
number
|
1
Min: 0.5 Max: 5 |
PAG scale for the second pass
|
| second_pass_guidance_scale |
number
|
4
Min: 1 Max: 14 |
Guidance scale for the second pass
|
| second_pass_outpaint_strength |
number
|
1
Max: 1 |
Outpaint strength for the second pass
|
| second_pass_depth_strength |
number
|
1
Max: 1 |
Depth strength for the second pass. We now want to lock in the depth, so 1.0 is a good starting point
|
| second_pass_line_strength |
number
|
1
Max: 1 |
Line strength for the second pass
|
| skip_light_pass |
boolean
|
False
|
Skip the light pass
|
| light_pass_steps |
integer
|
10
Min: 10 Max: 50 |
Number of steps for the light pass. This is a two step process and takes 1.9x steps
|
| light_pass_guidance_scale |
number
|
7
Min: 1 Max: 14 |
Guidance scale for the light pass
|
| light_mix_scale |
number
|
0.5
Max: 1 |
Mix scale for the light pass. 0.5 is a good starting point
|
Output schema
The shape of the response you’ll get when you run this model with an API.
{'items': {'format': 'uri', 'type': 'string'},
'title': 'Output',
'type': 'array',
'x-cog-array-type': 'iterator'}