Flux 1 SRPO
FLUX.1 SRPO [dev] is a 12 billion parameter flow transformer that generates high-quality images from text with incredible aesthetics. It is suitable for both personal and commercial use and is optimized for high-performance inference.
API Endpoint
POST https://image.siraya.pro/v1/images/generations
Request Parameters
Header
| Parameter | Type | Required | Description |
|---|---|---|---|
Authorization |
string | Yes | Your API Key (e.g., Bearer <API_KEY>). |
Body
| Parameter | Type | Default | Description |
|---|---|---|---|
model |
string | - | The model ID: flux-1-srpo-text-to-image. |
prompt |
string | - | A text description of the image you want to generate. |
size |
string | landscape_4_3 |
The size of the generated image. Options: square_hd, square, portrait_4_3, portrait_16_9, landscape_4_3, landscape_16_9. |
num_inference_steps |
integer | 28 |
The number of denoising steps for image generation. |
guidance_scale |
number | 4.5 |
Measures how closely the model sticks to the prompt. |
acceleration |
string | regular |
Speed of generation: none, regular, high. |
seed |
integer | - | Use a specific seed for deterministic generation. |
enable_safety_checker |
boolean | false |
Enable/disable content safety filtering. |
n |
integer | 1 |
Number of images to generate. |
output_format |
string | url |
The output format, typically url. |
[!TIP] For custom image sizes, you can pass an object to
image_sizewithwidthandheightproperties instead of using thesizepreset.
Usage Examples
curl https://image.siraya.pro/v1/images/generations \
-H "Content-Type: application/json" \
-H "Authorization: Bearer <API_KEY>" \
-d '{
"model": "flux-1-srpo-text-to-image",
"prompt": "A cute baby sea otter.",
"size": "landscape_4_3",
"num_inference_steps": 28,
"guidance_scale": 4.5,
"acceleration": "regular",
"n": 1,
"output_format": "url"
}'
import requests
url = "https://image.siraya.pro/v1/images/generations"
headers = {
"Content-Type": "application/json",
"Authorization": "Bearer <API_KEY>"
}
data = {
"model": "flux-1-srpo-text-to-image",
"prompt": "A cute baby sea otter.",
"size": "landscape_4_3",
"n": 1
}
response = requests.post(url, headers=headers, json=data)
print(response.json())