metadata

language:
  - en
license: creativeml-openrail-m
tags:
  - stable-diffusion
  - stable-diffusion-diffusers
  - text-to-image
  - diffusers
inference: true
pipeline_tag: text-to-image

Inference API Documentation

Overview

The Inference API allows you to make inference requests to perform image processing tasks using a remote service. This API supports various operations and requires specific input parameters.

Base URL

The base URL for the Inference API is: <hf_endpoint_url>

Authentication

The Inference API requires authentication using a bearer token. Include the token in the Authorization header of your requests.

Request Format

Send a POST request to the endpoint URL with the following JSON payload:

{
  "inputs": "<positive_prompt>",
  "negative_prompt": "<negative_prompt>",
  "height": <height>,
  "width": <width>,
  "guidance_scale": <guidance_scale>,
  "inference_steps" : <inference_steps>,
}

Request Parameters

Parameter	Type	Required	Description
inputs	string	Yes	The positive prompt for the inference.
negative_prompt	string	No	The negative prompt for the inference (optional).
height	integer	Yes	The height of the image.
width	integer	Yes	The width of the image.
guidance_scale	float	Yes	The guidance scale for the inference.
inference_steps	integer	No	The steps for inference.(25 default)

Response Format

The API response will be a JSON object with the following structure:

{
  "image": "<base64_encoded_image>"
}

Response Format

The API response will be a JSON object with the following structure:

Field	Type	Description
image	string	The base64-encoded image generated by the API.

Example Request

Here's an example request using Python:

import requests

url = '<hf_endpoint_url>'
token = '<hf_token>'

requestData = {
  'inputs': 'Positve prompt',
  'negative_prompt': 'Negative prompt goes here',
  'height': 512,
  'width': 512,
  'guidance_scale': 7.5,
  'inference_steps': 50,
}

headers = {
  'Authorization': 'Bearer ' + token,
  'Content-Type': 'application/json'
}

response = requests.post(url, json=requestData, headers=headers)
print(response.json())

Here's an example request using JavaScript:

const endpointURL = '<hf_endpoint_url>';
const hfToken = '<hf_token>';

const requestData = {
  
  s: 'Positve prompt',
  negative_prompt: 'Negative prompt goes here',
  height: 512,
  width: 512,
  guidance_scale: 7.5,
  inference_steps: 50,
};

const headers = {
  'Authorization': `Bearer ${hfToken}`,
  'Content-Type': 'application/json'
};

fetch(endpointURL, {
  method: 'POST',
  body: JSON.stringify(requestData),
  headers: headers
})
  .then(response => response.json())
  .then(data => console.log(data))
  .catch(error => console.error(error));