cjwbw/stable-diffusion-v2

sd-v2 with diffusers, test version!

Input

Configure the inputs for the AI model.

Seed

Random seed. Leave blank to randomize the seed

width

Width of output image. Maximum size is 1024x768 or 768x1024 because of memory limits

height

Height of output image. Maximum size is 1024x768 or 768x1024 because of memory limits

Prompt

Input prompt

scheduler

Choose a scheduler. Seems only DDIM and K_EULER and DPMSolverMultistep work for sd-v2 now.

Init Image

Initial image to generate variations of. Will be resized to the specified width and height

Num Outputs

Number of images to output. Currenly allowing 1-3, otherwise would OOM.

Guidance Scale

Scale for classifier-free guidance

Negative Prompt

The prompt NOT to guide the image generation. Ignored when not using guidance

Prompt Strength

Prompt strength when using init image. 1.0 corresponds to full destruction of information in init image

Num Inference Steps

500

Number of denoising steps

Output

The generated output will appear here.

Click "Generate" to create an output.