sdxl-1.0

by Stability AI Image Generation

Open-source workhorse. Massive ecosystem. ControlNet compatible. The most versatile image model available.

Up to 1024×1024 ~2s avg From $0.01/image ControlNet support
View API Docs →

Specifications

Model ID
sdxl-1.0
Provider
Stability AI
Type
Image Generation
Architecture
Latent Diffusion (Dual Text Encoder)
Max Resolution
1024×1024
Supported Sizes
512×512, 768×768, 1024×1024, 1024×768, 768×1024
Avg Generation Time
~2 seconds
ControlNet
Canny, Depth, Pose, Tile, IP-Adapter
Img2Img / Inpainting
Both supported
Negative Prompt / Seed
Both supported
Steps / Guidance
10–50 (def: 30) / 1.0–20.0 (def: 7.5)
Schedulers
DPM++ 2M Karras, Euler A, DDIM, UniPC
Output Formats
PNG, JPEG, WebP
Batch Generation
Up to 4 images per request
API Endpoint
POST /v1/images/generations
OpenAI Compatible
Yes — standard OpenAI SDK

Generated with sdxl-1.0

Try sdxl-1.0

sdxl-1.0

50 free generations. No credit card required.

API integration

OpenAI-compatible. Drop-in replacement.

from openai import OpenAI

client = OpenAI(
    base_url="https://api.fluxpool.ai/v1",
    api_key="fp_your_api_key"
)

response = client.images.generate(
    model="sdxl-1.0",
    prompt="Isometric RPG village with tavern and market stalls, vibrant pixel art",
    size="1024x1024"
)

print(response.data[0].url)

"Isometric RPG village with tavern and market stalls, vibrant pixel art"

Generated with sdxl-1.0