← Back to all generators

nvidia/sana

A fast image model with wide artistic range and resolutions up to 4096x4096

Capabilities

1:14:33:416:99:162:33:2Negative PromptSeed

Cost

Community model (estimated from hardware time)

Input Parameters

guidance_scalenumber

Classifier-free guidance scale

Default: 5min: 1, max: 20
heightinteger

Height of output image

Default: 1024
model_variantstring

Model variant. 1600M variants are slower but produce higher quality than 600M, 1024px variants are optimized for 1024x1024px images, 512px variants are optimized for 512x512px images, 'multilang' variants can be prompted in both English and Chinese

Default: "1600M-1024px"
1600M-1024px1600M-1024px-multilang1600M-512px600M-1024px-multilang600M-512px-multilang
negative_promptstring

Specify things to not see in the output

Default: ""
num_inference_stepsinteger

Number of denoising steps

Default: 18min: 1
pag_guidance_scalenumber

PAG Guidance scale

Default: 2min: 1, max: 20
promptstring

Input prompt

Default: "a cyberpunk cat with a neon sign that says "Sana""
seedinteger

Random seed. Leave blank to randomize the seed

widthinteger

Width of output image

Default: 1024
Version: c6b5d2b74599Updated: 7/25/2026238.9K runs