Stable Diffusion XL 1.0

imageopenrail3.5B

SDXL base model with 3.5B UNet backbone and OpenCLIP-ViT/G + CLIP-ViT/L text encoders. Refiner model adds 6.6B for higher detail. Widely adopted community standard.

Capabilities

Infrastructure
Fine-tuning
Image Generation
Inpainting
Negative prompt
Image-to-image

Details

Model IDstable-diffusion-xl-1.0
Provider Stability AI
Creator stability
Familystable-diffusion
Licenseopenrail
Statusactive
Typeimage
Input modalitiestext, image
Output modalitiesimage
Open weightYes
Parameters3.5B
Architectureunet
Max resolution1536x1536
Resolutions1024x1024, 1152x896, 896x1152
Aspect ratios—
Batch size—
Release date
Sourcecommunity

Family Comparison: stable-diffusion

ModelMax resolutionResolutionsAspect ratiosBatch sizePricing
Stable Diffusion 3.5 Large Turbo1536x10241024x1024, 1024x1536, 1536x1024———
Stable Diffusion 3.5 Medium1536x10241024x1024, 1024x1536, 1536x1024———
Stable Diffusion 31536x10241024x1024, 1024x1536, 1536x1024———
Stable Diffusion XL 1.01536x15361024x1024, 1152x896, 896x1152———
Stable Diffusion 3.5 Large2048x20481024x1024, 1024x1536, 1536x1024, 2048x20481:1, 3:2, 2:3, 4:3, 3:4, 16:9, 9:16—$1.5$8

API

GET/v1/models/stability/stable-diffusion-xl-1.0

Stable Diffusion XL 1.0

stable-diffusion-xl-1.0

SDXL base model with 3.5B UNet backbone and OpenCLIP-ViT/G + CLIP-ViT/L text encoders. Refiner model adds 6.6B for higher detail. Widely adopted community standard.

Changes · 1 entries
stable-diffusion-xl-1.0create48ba559May 7, 2026, 02:17 PM