dazaleas / sd15-cog

Inference SD 1.5 with cog including several models.

  • Public
  • 1.8K runs
  • A100 (80GB)
Iterate in playground
Run with an API

Input

string

The model to use

Default: "pastelmix.safetensors"

string

The vae to use

Default: "None"

string
Shift + Return to add a new line

The prompt

Default: "masterpiece, illustration, portrait of a beautiful young woman"

string
Shift + Return to add a new line

The negative prompt (For things you don't want)

Default: "low quality, bad quality, worst quality"

string

The sampler used when generating

Default: "DPM++ 2M Karras"

integer
(minimum: 1, maximum: 100)

The steps when generating

Default: 35

number
(minimum: 1, maximum: 20)

CFG Scale defines how much attention the model pays to the prompt when generating

Default: 7

integer
(minimum: 1, maximum: 2048)

The width of the image

Default: 768

integer
(minimum: 1, maximum: 2048)

The height of the image

Default: 768

integer
(minimum: 1, maximum: 4)

Number of images to generate (1-4)

Default: 1

boolean

Generate high resoultion version

Default: false

integer
(minimum: 1, maximum: 4)

The scale to resize

Default: 2

number
(minimum: 0, maximum: 1)

The strength when applying denoising

Default: 0.35

integer
(minimum: 0, maximum: 30)

The steps when performing second pass

Default: 18

string

The upscaler to use when performing second pass

Default: "R-ESRGAN 4x+ Anime6B"

integer

The seed used when generating, set to -1 for random seed

Default: -1

Output

output
Generated in

This output was created using a different version of the model, dazaleas/sd15-cog:b3090867.

Run time and cost

This model costs approximately $0.038 to run on Replicate, or 26 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia A100 (80GB) GPU hardware. Predictions typically complete within 28 seconds. The predict time for this model varies significantly based on the inputs.

Readme

This model doesn't have a readme.