Stable-diffusion-xl-base-v10

Image Generation SDXL API

The Image Generation or Text-To-Image API endpoint allows you to generate an image from a textual description (prompt).

Introduction

Image Generation refers to the process of automatically generating visual content (such as images) based on textual descriptions provided as input.

AI Endpoints makes it easy, with ready-to-use inference APIs. Discover how to use them:

Model concept and configuration

This Image Generation API is based on an Open-Source model (Stable Diffusion XL): stabilityai/stable-diffusion-xl-base-1.0. It takes text as input (the prompt) and returns the corresponding image in color.

Model configuration:

• Input types: Raw text
• Language Support: English
• Output type: 1024x1024 image (bytes)

How to?

The Stable Diffusion XL (SDXL) endpoint offers you an optimized way to generate an image from:

• a prompt, which is a textual description or instruction that is used to guide the model in creating a new image
• a negative prompt (optional), which is a directive given to the model to specify what should not be present in the generated image

Learn how to use them with the following examples:

With a cURL

Consider changing the your-access-token by your token value before running the following command:

curl -X POST "https://stable-diffusion-xl.endpoints.kepler.ai.cloud.ovh.net/api/text2image" \
    -H 'accept: application/octet-stream' \
    -H 'content-type: application/json' \
    -H 'Authorization: Bearer <your-access-token>' \
    -d '{"negative_prompt":"Ugly, unrealistic","prompt":"Photo of a hippopotamus dressed suit and tie sitting in a coffee shop with a matcha latte, award winning photography, Elke vogelsang, 4k"}' \
    --output -

Warning:

• Note that you will get the output image as "bytes" by using --output -
• Consider adding an image filename in the command if you want to save it locally: --output my_image.jpeg

With a simple HTTP client (requests)

First install the requests library and pillow dependencies:

pip install requests pillow

Next, export your access token to the OVH_AI_ENDPOINTS_ACCESS_TOKEN environment variable:

export OVH_AI_ENDPOINTS_ACCESS_TOKEN=<your-access-token>

If you do not have an access token key yet, follow the instructions in the AI Endpoints – Getting Started.

Finally, run the following python code:

import os
import io
import requests
import json
from PIL import Image

url = "https://stable-diffusion-xl.endpoints.kepler.ai.cloud.ovh.net/api/text2image"

data = {
    "prompt": "Photo of a hippopotamus dressed suit and tie sitting in a coffee shop with a matcha latte, award winning photography, Elke vogelsang, 4k",
    "negative_prompt": "Ugly, unrealistic",
}

headers = {
    "accept": "application/octet-stream",
    "content-type": "application/json",
    "Authorization": f"Bearer {os.getenv('OVH_AI_ENDPOINTS_ACCESS_TOKEN')}",
}

response = requests.post(url, headers=headers, data=json.dumps(data))
if response.status_code == 200:
    # Handle response
    response_data = response.content
    img = Image.open(io.BytesIO(response_data))
    img.show()
else:
    print("Error:", response.status_code)

You will obtain the following (1024x1024) image:

Model rate limit

When using AI Endpoints, the following rate limits apply:

• Anonymous: 2 requests per minute, per IP and per model.
• Authenticated with an API access key: 400 requests per minute, per Public Cloud project and per model.

If you exceed this limit, a 429 error code will be returned.

If you require higher usage, please get in touch with us to discuss increasing your rate limits.