AI Endpoints: catalogue and playground
Compare and test open-weight models for text, code, voice and vision, no sign-up required. Then move to production with the same OpenAI-compatible API.
stable-diffusion-xl-base-v10
Image GenerationSDXL, developed by Stability AI, is an advanced text-to-image model released in July 2023, offering enhanced image generation capabilities.
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 $OVH_AI_ENDPOINTS_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 my_image.jpegNote:
- By default the command above save the image directly, as my_image.jpeg. To obtain the raw image bytes, use --output -
With a cURL, using our OpenAI compatible endpoint
Consider changing the your-access-token by your token value before running the following command:
curl -X POST "https://oai.endpoints.kepler.ai.cloud.ovh.net/v1/images/generations" \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $OVH_AI_ENDPOINTS_ACCESS_TOKEN" \
-d '{"model":"stabilityai/stable-diffusion-xl-base-1.0", "prompt":"Photo of a hippopotamus dressed suit and tie sitting in a coffee shop with a matcha latte, award winning photography, Elke vogelsang, 4k"}'Note: This endpoint returns a JSON payload with a b64_json field containing the base64 encoded image and does not accept negative_prompt.
With a simple HTTP client (requests)
First install the requests library and pillow dependencies:
pip install requests pillowNext, 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 base64
import requests
from PIL import Image
# You can use the model dedicated URL
url = "https://stable-diffusion-xl.endpoints.kepler.ai.cloud.ovh.net/api/text2image"
# Or our unified endpoint for easy model switching with optimal OpenAI compatibility
url = "https://oai.endpoints.kepler.ai.cloud.ovh.net/v1/images/generations"
# Set prompts (negative prompt is not used when using OpenAI compatibility)
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 = {
"Authorization": f"Bearer {os.getenv('OVH_AI_ENDPOINTS_ACCESS_TOKEN')}",
}
if "oai.endpoints" in url:
# OpenAI-compatible endpoint
headers["Content-Type"] = "application/json"
payload = {
"model": "stable-diffusion-xl-base-v10",
"prompt": prompt,
"size": "1024x1024",
}
response = requests.post(url, headers=headers, json=payload)
if response.status_code == 200:
resp_json = response.json()
img_b64 = resp_json["data"][0]["b64_json"]
img_bytes = base64.b64decode(img_b64)
img = Image.open(io.BytesIO(img_bytes))
img.show()
else:
print("Error:", response.status_code, response.text)
else:
# Raw endpoint
headers["accept"] = "application/octet-stream"
headers["content-type"] = "application/json"
payload = {
"prompt": prompt,
"negative_prompt": negative_prompt,
}
response = requests.post(url, headers=headers, json=payload)
if response.status_code == 200:
img = Image.open(io.BytesIO(response.content))
img.show()
else:
print("Error:", response.status_code, response.text)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.
Going Further
For a broader overview of AI Endpoints, explore the full AI Endpoints Documentation.
Reach out to our support team or join the OVHcloud Discord #ai-endpoints channel to share your questions, feedback, and suggestions for improving the service, to the team and the community.
Get started with Stable-diffusion-xl-base-v10
AI Endpoints - Getting Started
New to AI Endpoints? This guide walks you through everything you need to get an access token, call AI models, and integrate AI APIs into your apps with ease.
Start TutorialAI Endpoints - Features, Capabilities and Limitations
Explore what AI Endpoints can do. This guide breaks down current features, future roadmap items, and the platform's core capabilities so you know exactly what to expect.
Start TutorialAI Endpoints - Troubleshooting
Running into issues? This guide helps you solve common problems on AI Endpoints, from error codes to unexpected responses. Get quick answers, clear fixes, and helpful tips to keep your…
Start Tutorial