Text To Image

[POST] https://api.nichetensor.com/api/v1/txt2img

Header

NAME	TYPE	REQUIRED
`API_KEY`	string	✔️

Body

NAME	TYPE	REQUIRED	DESCRIPTION	Available Choices
`prompt`	string	✔️	The textual prompt used to generate the image.
`model_name`	string	✔️	The name of the model to use for image generation.	`AnimeV3`, `DreamShaperXL`, `JuggernautXL` , `RealitiesEdgeXL`, `StickerMaker`
`seed`	integer		The seed value for the random number generator. Use `0` for a random seed. Default: `0`
`aspect_ratio`	string		Ratio of image to generate. Default: `1:1`	`1:1`, `2:3`, `3:2` , `16:9`, `9:16`, `13:19`, `19:13`
`negative_prompt`	string		List things to avoid to generation. Empty by default
`advanced_params`	object		Advanced params such as number of steps and guidance scale.
`advanced_params.use_expansion`	bool		Use prompt expansion for more aesthetic output	`True`, `False`

Example Request in Python

import requests
 
API_KEY = "YOUR_API_KEY" 
endpoint = "https://api.nichetensor.com/api/v1/txt2img"
headers = {"API_KEY": API_KEY}

data = {
    "prompt": "a cute cat",
    "model_name": "AnimeV3",
    "seed": 0,
    "aspect_ratio": "13:19",
    "negative_prompt": "low quality",
    "advanced_params": {"num_inference_steps": 20, "use_expansion": True}
}
 
response = requests.post(endpoint, json=data, headers=headers)
 
print(response.status_code)
print(response.json())

Example Response

{ // image is base64-encoded
		"image": "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRofHh0aHBwgJC4n" # base64 image
}

The base64-encoded image can be converted to PIL using:

def base64_to_pil_image(base64_image):
    image = base64.b64decode(base64_image)
    image = io.BytesIO(image)
    image = Image.open(image)
    return image