Stable Diffusion's newest model. Generates a wide variety of images reflecting different styles without complex prompting.
Approx. Price
$0.111 per image
Model Type
image
Settings
Generation controls available for this model.
Images Per Run
Up to 4
Output images
Input Images
N/A
No reference/edit image input
Output Sizes
Guidance Scale
Default
3.5
How closely to follow the prompt. Higher values stick closer to the prompt but may be less creative.
Negative Prompt
Default
N/A
Details you don't want in the image (e.g. blurry, low quality, etc)
Number of Images
Default
1
Resolution
Default
1024x1024
Options (5)
1024x1024 (Square (1:1)), 1024x768 (Landscape (4:3)), 1024x576 (Landscape (16:9)), 768x1024 (Portrait (3:4)) +1 more
Steps
Default
28
Number of denoising steps. Higher values may improve quality but take longer.
Benchmarks
Benchmarks
Human preference benchmarks sourced from LMArena.
Text to Image
#74 / 75
Arena Score
938.3
Votes
23,396
Confidence Interval
933.8 - 942.8
Published 2026-08-07 · Matched as stable-diffusion-v35-large
LMArena DatasetExamples
Loading examples…
Related image models
Compare SD 3.5 Large with similar models from the same provider or model family.
SD 3.5 Large Turbo
stable-diffusion-v35-large/turboTurbo version of Stable Diffusion's newest model. Faster and cheaper performance while still maintaining great prompt adherence and quality.
Stable Diffusion XL
fast-sdxlCheap and powerful text-to-image model that generates pictures rapidly.
Stable Diffusion 3 Medium
sd3_base_medium.safetensorsExcels at photorealism, typography, and prompt following. Works best in 1024x1024.
Fluently XL V3 Lightning
stable-diffusion-xl-turboFluently XL V3 Lightning - Fast, high-quality image generation. CivitAI model 324891@408626.
Krea 2 Large
wavespeed-ai/krea-v2-large/text-to-imageKrea 2 Large generates high-fidelity images from text prompts with aspect ratio, creativity, and optional style reference controls.
Krea 2 Large
krea/v2/large/text-to-imageKrea 2 Large generates high-fidelity images with strong aesthetic control, tunable creativity, and optional style references.