bytedance / sdxl-lightning-4step

SDXL-Lightning by ByteDance: a fast text-to-image model that makes high-quality images in 4 steps

  • Public
  • 826.4M runs
  • A100 (80GB)
  • GitHub
  • Weights
  • Paper
  • License
Run with an API
Playground API Examples README Versions

Run bytedance/sdxl-lightning-4step with an API

Input schema

seedinteger

Random seed. Leave blank to randomize the seed

widthinteger

Width of output image. Recommended 1024 or 1280

Default
1024
Minimum
256
Maximum
1280
heightinteger

Height of output image. Recommended 1024 or 1280

Default
1024
Minimum
256
Maximum
1280
promptstring

Input prompt

Default
"self-portrait of a woman, lightning in the background"
schedulerstring

scheduler

Default
"K_EULER"
num_outputsinteger

Number of images to output.

Default
1
Minimum
1
Maximum
4
guidance_scalenumber

Scale for classifier-free guidance

Maximum
50
negative_promptstring

Negative Input prompt

Default
"worst quality, low quality"
num_inference_stepsinteger

Number of denoising steps. 4 for best results

Default
4
Minimum
1
Maximum
10
disable_safety_checkerboolean

Disable safety checker for generated images

Output schema

Type
uri[]