You're looking at a specific version of this model. Jump to the model overview.
ultracoderru /nova-anime-xl-17:8f702486
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 |
|---|---|---|---|
| prompt |
string
|
1girl, solo, cute
|
Positive prompt. Put here a comma-separated list of Danbooru tags to describe, what should be present on image(s). You can add weights to tags and use BREAK token.
See https://civitai.com/articles/19107/crodys-illustrious-noobai-image-generation-tips#2.-prompting-guide-and-resources
|
| negative_prompt |
string
|
|
Negative prompt. Put here a comma-separated list of Danbooru tags to describe, what should NOT be present on image(s). Syntax is the same as for positive prompt.
|
| add_recommended_tags |
boolean
|
True
|
If true, the recommended tags will be added to both prompts. In that case you only need to describe a scene.
|
| width |
integer
|
1024
|
Width of output image(s). It is recommended to pick resolutions where total pixels remain under 1-megapixel threshold.
|
| height |
integer
|
1024
|
Height of output image(s). It is recommended to pick resolutions where total pixels remain under 1-megapixel threshold.
|
| num_outputs |
integer
|
1
Min: 1 Max: 10 |
Number of images to generate
|
| num_inference_steps |
integer
|
20
Min: 1 Max: 100 |
Number of denoising steps
|
| guidance_scale |
number
|
4.5
Min: 1 Max: 20 |
Scale for classifier-free guidance (CFG). It controls how much the image generation follows the prompt. Higher values make the output more closely aligned with the prompt, but may reduce image quality. Lower values allow more creative freedom, but may stray from the prompt.
|
| scheduler |
None
|
Euler a
|
Scheduler to use ('Euler a' is recommended)
|
| seed |
integer
|
-1
|
Random seed. Set to -1 to pick random seed.
|
| hires_enable |
boolean
|
False
|
Upscale the image(s) (apply Hires.fix).
|
| hires_factor |
number
|
1.5
Min: 1 Max: 4 |
Upscaling factor
|
| hires_strength |
number
|
0.4
Max: 1 |
Denoising strength for the upscaling stage. It controls how much the upscaler can change the image(s).
|
| hires_num_inference_steps |
integer
|
40
Min: 1 Max: 100 |
Number of denoising steps for the upscaling stage
|
| hires_guidance_scale |
number
|
4
Min: 1 Max: 20 |
Scale for classifier-free guidance (CFG) for the upscaling stage
|
| disable_safety_checker |
boolean
|
False
|
Disable safety checker for generated images. This feature is only available through the API.
|
| adjust_brightness |
array
|
Brightness adjustments to apply to the image(s) [expert setting]
The adjustments will be applied after low-resolution generation. The value must be a list of THREE float numbers - adjustment factors for shadows, midtones and highlights. Numbers must be between 0 and 10, 1.0 means no adjustment.
|
|
| adjust_contrast |
array
|
Contrast adjustments to apply to the image(s) [expert setting]
Rules are the same as for brightness adjustments.
|
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'}