You're looking at a specific version of this model. Jump to the model overview.
sergeishapovalov /refaopt:3f028cd8
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
|
Prompt
|
|
| width |
integer
|
0
Max: 1280 |
Width of output image
|
| height |
integer
|
0
Max: 1280 |
Height of output image
|
| num_outputs |
integer
|
1
Min: 1 Max: 4 |
Number of images to output
|
| sampler |
None
|
Euler
|
Sampling method для Flux моделей
|
| scheduler |
None
|
Simple
|
Schedule type для Flux моделей
|
| num_inference_steps |
integer
|
8
Min: 1 Max: 200 |
Number of denoising steps
|
| guidance_scale |
number
|
1
Min: 1 Max: 50 |
CFG Scale (для Flux рекомендуется значение 1.0)
|
| distilled_guidance_scale |
number
|
3.5
Max: 30 |
Distilled CFG Scale (основной параметр для Flux, рекомендуется 3.5)
|
| seed |
integer
|
-1
|
Random seed. Leave blank to randomize the seed
|
| enable_hr |
boolean
|
True
|
Hires. fix
|
| hr_upscaler |
None
|
4x-UltraSharp
|
Upscaler for Hires. fix
|
| hr_steps |
integer
|
8
Max: 100 |
Inference steps for Hires. fix
|
| hr_scale |
number
|
1.3
Min: 1 Max: 4 |
Factor to scale image by
|
| denoising_strength |
number
|
0.3
Max: 1 |
Denoising strength. 1.0 corresponds to full destruction of information in init image
|
| debug_flux_checkpoint_url |
string
|
|
Flux checkpoint URL. UPDATE 18.05.2025 BREAKS IT BECAUSE OF NEW def calculate_shorthash(self):
|
| enable_clip_l |
boolean
|
False
|
Enable encoder
|
| enable_t5xxl_fp16 |
boolean
|
False
|
t5xxl_fp16
|
| enable_ae |
boolean
|
False
|
Enable ae
|
| force_model_reload |
boolean
|
False
|
Load Flux model from scratch
|
| forge_unet_storage_dtype |
None
|
Automatic (fp16 LoRA)
|
forge_unet_storage_dtype
|
| image |
string
|
|
Input image for image to image mode. The aspect ratio of your output will match this image
|
| prompt_strength |
number
|
0.8
Max: 1 |
Prompt strength (or denoising strength) when using image to image. 1.0 corresponds to full destruction of information in image.
|
| aspect_ratio |
None
|
9:16
|
Aspect ratio for the generated image
|
| output_format |
None
|
webp
|
Format of the output images
|
| lora_urls |
array
|
[]
|
Ссылки на LoRA файлы
|
| lora_scales |
array
|
[1]
|
Lora scales
|
| ad_prompt |
string
|
|
None
|
| ad_hands_prompt |
string
|
|
None
|
| postback_url |
string
|
|
None
|
| adetailer |
boolean
|
False
|
Enable adetailer
|
| adetailer_args |
string
|
{}
|
Adetailer arguments
|
| adetailer_args_hands |
string
|
{}
|
Adetailer arguments for hands
|
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'}